the history of the development and operation of information...
TRANSCRIPT
The history of the development and operation of
Information Technology at the Potchefstroom Campus of
the PU for CHE from approximately 1967 to 2003
Cobus van der Walt
Copyright © 2018 by Cobus van der Walt
All rights reserved. This book or any portion thereof may not
be reproduced or used in any manner whatsoever without
the express written permission of the publisher except
for the use of brief quotations in a book review.
In association with
NWU: Information Technology.
Images used under the license from GALLO IMAGES / GETTY IMAGES
Editor:
Professor Annette Combrink
Design and layout:
Graphic Services, North-West University, South Africa
ISBN:
?
Printed in the Republic of South Africa
First Edition, 2018
02 03 04 05 06 07 08 09 10 11 - 26 25 24 23 22 21 20 19 18 17
CONTENT
1. Introduction
2. The Beginning
3. The IBM 370/125 from 1974 to 1980
4. The 1980s (1981-1990)
5. The coordinating role of the Committee
for Telecommunication Networks
6. Library computerisation
7. Information Technology and
Management, 1991 to 2003
1
5
19
33
71
81
87
H
T
PREFACE istorical writing is, by nature, selective and cannot convey all information
in detail. The same applies to this document. There is an abundance of
meeting minutes and reports available from which only single core decisions
could be conveyed. There are also many photos available on certain topics,
but only a limited number of them could be used. On the contrary, some
topics under discussion have very few or no photographs available at all.
Contributions in this regard will be much appreciated and valued, as will
photos of staff members who do not appear on any of the photos in this
document. In addition, any comments on factual inaccuracies and missing
core information will be valued.
he documents referred to in this document, together with any other
related documents, have been saved on a CD titled “IT history PU for
CHE 1967-2003” that is kept in the NWU archive. The documents in Winword
format that are referred to can be found on the same CD in the IT history\
Winword\ directory.
1
T
CHAPTER
Introduction
oday, it is difficult to imagine a world
without the variety of computers,
including desktops, laptops, tablets,
smartphones and the unseen
computers in most of our household
appliances (and much more) that we
take for granted. Computers have an
immense impact on our daily lives and
have changed the nature of scientific
research completely. This has, however,
not always been the case and many of
us grew up with only a vague notion
that somewhere out there existed
something called computers that had
basically no visible impact on our daily
lives.
1
Although numerical calculations have
always been part of most research in
disciplines like Statistics and Physics,
the execution thereof was often
laborious and time-consuming. For
calculations that did not require
extreme accuracy, the slide rule was
a solution, but for more accurate
work, manual and electric calculators
had to be used. This process was
quite tedious and required thorough
checks to prevent input errors. As such,
there was an early demand for fast,
accurate electronic computers. In some
instances, calculations were complex,
e.g. differential equations, which often
required ingenious solutions. Prof PH
Stoker mentioned that, on suggestion
of Pierre Haarhoff, a post-graduate
student, integrals of intricate functions
were calculated by drawing the graphs
on paper, cutting it out accurately and
then weighing it1.
A serious demand developed for
electronic computing facilities at the
University. It thus comes as no surprise
that the purchase of the first computer
was strongly advocated for.
Schematic representation of the main hardware components used for calculations at the PU for CHE from the 1950s to the present time, from the slide rule to mechanical calculators,, to electrical calculators, to printing calculators, to the IBM 1130 and the mainframe era, to the first powerful servers like the IBM RS/6000, and presently the modern compact servers with multiple hard drives
1 Transcription of the Author’s interview with Prof PH Stoker, 7 September 2009
2 3
2
CHAPTER
The Beginning
EARLY HISTORY2
The University was, early on, aware of
the possible advantages of a computer
for mathematical sciences research in
particular. Contact was established with
the important computer companies in
South Africa at the time, like IBM South
Africa and ICT. In approximately 1959,
IBM South Africa invited the University
to inspect a new computer that Massey-
Ferguson had installed in Vereeniging.
A large delegation of staff members
paid a visit to this new computer, but
nothing further transpired17.
In about 1961, Dr FJ du Plessis from
Economic Sciences commissioned
the Department of Instrument Making
to build what turned out to be an
interesting example of, in essence, an
analogue computer. This computer
could simulate the effect of external
factors on the economy. Money was
represented by a coloured liquid, the
flow of which was regulated by control
valves, pumps and feedback channels.
2 Based, for the most part, on transcriptions of interviews the Author conducted with Prof. JF de Beer, Prof. T Steyn and Mrs K Benadie on 19 November 2008 and Mrs A Oosthuyzen on 20 November 2017.
5
6 7
In the 1960s, a Johannesburg company
demonstrated a commercial analogue
computer on campus. The problems
that could be solved with this computer
were limited and the demonstration
generated very little interest.
During this time, Mr (later Dr) JAM
de Villiers, who was studying towards
his PhD in Physics (on the scattering
of muons), had to perform complex
calculations for the theoretical curves
that he had to compare with his
experimental values. These calculations
were performed for him by Mr CE
Harding on the ZEBRA computer in
STC’s computer department3. STC
later wanted to donate the ZEBRA
computer to the University, but
the donation was impractical and,
subsequently, rejected. The reason
for this rejection was that ZEBRA
computers used vacuum tubes, which
took up a lot of space and required a
lot of maintenance. All programming
on the ZEBRA was done in Assembler
and not a high-level language, which
did not make it very user-friendly. In
addition, the arrival of more modern
and compact computers that made use
of semiconductors, made the ZEBRA a
less attractive option.
Dr (later Prof) JF de Beer, in partnership
with Dr (later Prof) JH Venter from
Statistics, manually calculated
the University obtaining a computer.
With the (possible) help of Prof AJE
Sorgdrager from Economic Sciences,
the matter was taken up with the rector
at the time, Prof HJJ Bingle. In the early
1960s, during his doctoral studies at the
University of Chicago, Dr Venter gained
experience working with Fortran on the
IBM 7090 of the University in question4.
In 1965/66, during a research visit to
the University of Durham, England,
Dr de Beer, on his part, gained
extensive experience in solving physics
calculation problems with a computer.
Both Dr Venter and Dr de Beer were
eager to use a computer. They were
supported by Prof WP Robbertse, the
Dean of Natural Sciences at the time5.
In approximately 1966, the University of
the Witwatersrand wanted to sell their
IBM 1620 computer and the PU for CHE
sent a delegation there to investigate
the matter. At the time, the PU staff
were already aware of the IBM 1130,
the IBM 1620’s successor, and decided
against the purchase. The process was
then started to obtain approval for the
purchase of a new IBM 1130 computer.
THE PURCHASE OF THE
FIRST COMPUTER
Prof JM de Wet from Statistics
presented a formal proposal to
on 31 August 19666. Prof De Wet was
supported by Prof WP Robbertse
from Mathematics and Applied
Mathematics, who also served as the
Dean of Natural Sciences at the time,
and Prof PH Stoker from Physics. Prof
De Wet’s thorough and well-argued
proposal was accepted by the Finance
Committee on 27 September 19667 and
confirmed by the University Council on
28 September 19668.
The IBM 1130 was ordered and
expected to be delivered in about one
year. The total purchase price of the
computer was R37 5009, approximately
R2.6 million in 2017 Rand (ZAR) value. In
the interim, a card punch was purchased
to capture data and programmes on
punch cards for use on computers in
the vicinity.
PROGRAMMING
LANGUAGE TRAINING
During October 1966, Dr JF de Beer
presented a course in Fortran
programming for interested parties.
Several scientists from Physics,
Statistics and Mathematics attended
the course. At the end of the course,
the attendees’ programming attempts
were tested on the IBM 360 computer
of the Atomic Energy Board at
Pelindaba. For the researchers, this
visit and their subsequent expertise
in Fortran programming emphasised
the power of the computer in solving
calculation problems, and thus the
arrival of the University’s first computer
was awaited with great anticipation.
From time to time, Physics researchers
in particular had to use computers in
the vicinity, like those at the Atomic
Energy Board and the Universities of
Pretoria and the Witwatersrand. This
further emphasised the urgent need for
the IBM 1130.
Before the arrival of the IBM 1130, IBM
South Africa presented a week-long
course in Assembler for University
staff. This training was attended by Dr
JF de Beer, Dr JH Venter and Dr DJ
van Wyk10.
Figure 1 Prof JF (Hannes) de Beer
simulations of the penetrability of the encourage the purchase of an IBM
atmosphere by cosmic rays. It was an
exceptionally time-consuming exercise,
which emphasised the necessity of
1130 computer (a matter which he
referred to as exceptionally urgent) to
the Faculty Board of Natural Sciences
6 Faculty Board of Natural Sciences, minutes, 31 August 1966, Art 33 7 Finance Committee, minutes, 27 September 1966, Art 21 8 University Council, minutes, 28 September 1966, Art 9 XIX 9 According to Prof JH Venter’s recollection, the IBM 1130 was made available to
the University at a special price, through the mediation of Mr JF Clarke, the Sales 3 JAM de Villiers, D.Sc. thesis, PU for CHE, 1961 4 Transcription of Author’s interview with Prof. JH Venter on 2 September 2009 5 Personal conversation between the Author and Prof WP Robbertse in 1996.
Manager of IBM South Africa. 10 Giel Hattingh, Rekenaarwetenskap en Inligtingstelsels, Chapter 4.6 in ’n Bekroonde
Nalatenskap: Die Fakulteit Natuurwetenskappe oor 100 jaar, Editor: Elize S van Eeden, 2017
8 9
PUTTING THE IBM 1130
INTO OPERATION
It was decided that Dr JF de Beer, at
the time a member of the Department
of Physics, would take the lead in
the computer centre that would be
established. In 1967, Ms Klaasje Moraal
(later Mrs Klaasje Benadie) was the first
full-time employee appointed at the
centre.
The IBM 1130 was delivered in July
1967 (earlier than anticipated) when Dr
De Beer was abroad. The University did
not own a forklift to move the heavy
computer. Mr MGL Bosman from the
Department of Maintenance and his
team moved the computer, which was
still in its packaging, from the sidewalk
to the Physics building’s elevator by
rolling it on pipes. It was then moved
to the second floor (via the elevator),
where it was, again, rolled on pipes to
the computer room.
Figure 2 The University’s IBM 1130 computer in 1971. The computer is in the foreground and the card reader/punch is in the background. The
This computer room was previously
used by Statistics students and third-
year Physics students for practical
sessions. A few changes were made
to the rooms to make them fit for use
as a computer centre. A false floor was,
however, not installed and the cables
rested on the floor.
The computer was set up by one of
IBM South Africa’s computer experts,
Mr Fred Macaskill, who was a PU for
CHE alumnus.
The IBM 1130 computer had a memory
of 32 kilobytes and a removable disk
drive of 1 megabyte1. Programmes
were uploaded to the computer
with a punch card reader, which also
functioned as a card punch. The punch
card reader or mechanical paper tape
reader was used for data input. The IBM
1130’s operating system was stored
on punched cards; a large pack of
cards had to be read into the machine
whenever the operating system had to
be reloaded. Output was generated
with a typewriter or a card punch
machine. In practice, it appeared that
the typewriter was unable to handle
Figure 3 The IBM 1130’s IBM 2310 hard drive. It
the workload, resulting in numerous
problems. The problem was only
solved when, after quite some time, a
line printer was purchased.
To demonstrate the computer’s
capability to interested parties, two
post-graduate Physics students, Mr JP
Maree and Mr DJ Joubert, respectively,
developed a programme that could
calculate the day of the week of any
date in history and one that could play
noughts and crosses with the computer.
The computer was officially handed
over to the University’s rector, Prof HJJ
Bingle, on 14 September 1967 by Mr JF
Clarke, IBM South Africa’s CEO. During
the ceremony, the photo in Figure 4 was
taken. Reports on the event appeared
in Die Wapad of 20 September 196711
and Die Veteraan of Sept/Nov 196712.
Figure 5 The people who were involved in the initial acquisition of the IBM 1130 computer, were present at the 25-year commemoration of the partnership between IBM/ISM and the University on 24 August 1992. They are (from left to right), Prof WP Robbertse, former Dean of Natural Sciences; Mr Wiid du Toit, former IBM South Africa salesperson; Mrs Klaasje Benadie, formerly Ms Klaasje Moraal, the computer centre’s first full-time staff member; Prof JF de Beer, the first Head of the Computer Centre; and Mr JF Clarke, former IBM South Africa CEO.
Figure 4 The official handover of the IBM 1130 to the Rector of the PU for CHE, Prof HJJ Bingle, (left) by the CEO of IBM South Africa, Mr JF Clarke (second from the left). Also present on the photo are (from left to right) Ms Klaasje Moraal, Dr JF de Beer and Prof WP Robbertse.
mechanical paper tape reader is on the far right of the photo. The identities of the two persons
had a capacity of 1 megabyte and a diameter of approximately 38 cm. (Source: Author’s photo
at the card reader are unknown. (Source: Photo collection, NWU archive)
collection) 11 Die Wapad: Vol 23 Nr 1, p5, 20 September 1967, accessed at http://dspace.nwu.
ac.za/bitstream/handle/10394/14016/Wapad_d23_n01_1967.pdf?sequence=1 12 Die Veteraan, p47, September/November 1967
10 11
THE OPERATION OF
THE IBM 1130
The IBM 1130 was a user-friendly
computer that could be operated by
users and did not require operating
staff. As such, initially, there were no
operating staff. The departmental
typists were responsible for completing
the scientists’ punch work.
The IBM 1130 was, almost since its
arrival, in full-time use, especially by
Statistics and Physics. Very soon, it
was also used continuously (including
after hours), especially by Mr (later
Dr) WF Coetzee, who conducted
calculation-intensive research into
the trajectories of cosmic ray particles
in the complex geomagnetic field.
Thereafter, similar work was conducted
from approximately 1969 to 1972 by
Mr (later Dr) PJ König’s for his doctoral
research project in Physics.
The IBM 1130 was later expanded to
include an IBM 1132 line printer to
replace the overworked typewriter,
and also two disk drives and two
magnetic tape drives. The memory
was also increased to 32 kilobytes.
Between 1970 and 1971, a second IBM
1130 was purchased to help with the
increasing workload. In August 1971,
a small graphic plotter was added13.
Before the plotter was obtained, rough
graphs could only be printed with the
line printer. This was done by changing
the position of a character in different
rows. This technique was developed by
Prof JH (Hennie) Venter.
The Department of Physics used
punched paper tape to process cosmic
ray data received from Antarctica
via telex transmission. The neutron
monitor results from Hermanus were
also stored on punched tape14. The IBM
1130’s mechanical paper tape reader
was very slow at 15 characters per
second. This resulted in bottlenecks,
since the IBM 1130 could only work on
one task at a time. In approximately
Figure 7 Mr (later Dr) JJ (Johan) van Loggerenberg (Source: Photo collection of
1971/1972, Prof de Beer purchased a
Facit optical paper tape reader, which
operated much faster at 500 characters
per second. There was no standard
interface for connecting the Facit
paper tape reader to the IBM 1130.
Prof de Beer approached Mr (later Dr)
AJ van der Walt of Electronic Services
to design and build a suitable interface.
This interface was designed with TTL-
integrated circuits. The programming
was done in Assembler by Prof de
Beer and Mr van der Walt, and the
Facit tape reader was successfully
commisioned. This was the University’s
first successful attempt to connect non-
IBM equipment to an IBM computer.
STAFF
As mentioned earlier, Dr JF de Beer
was the first head of the Computer
Centre. He was assisted by Ms Klaasje
Moraal, who was appointed in July
1967 as the centre’s first full-time staff
member.
The staff complement was later
extended and in 1969, Ms Dicky Otten
(later Mrs Dicky Venter) was appointed
as Scientific Programmer. During the
same time, Mrs Bettie Venter was
appointed as Punch Operator. She
did, however, quite soon take over
the operation of the salary system
(see below), as it took up a lot of
time. Mr (later Prof) Piet Bruwer joined
the team in 1971 or 1972; he was
Figure 6 The IBM 1132 line printer that was later purchased. (Source: Photo collection of Mr L Smuts)
JJ van Loggerenberg)
13 Die Wapad, Vol 26, Nr 6, p9, 7 May 1971, accessed at http://dspace. nwu.ac.za/bitstream/handle/10394/14430/Wapad_d26_n06_1971. pdf?sequence=1&isAllowed=y
14 The telex signals were transmitted via high-frequency radio from Antarctica to South Africa and contained an abundance of errors. It was common practice to transfer the data twice and then compare the two tapes’ data to identify and correct transmission errors.
12 13
Figure 8 Mr (later Prof) T (Tjaart) Steyn (Source: Photo collection of T Steyn)
Figure 9 Prof PJS Bruwer (left), Prof JF de Beer (right) and Prof PH Stoker (middle) were pioneers in the early years of the Computer Centre. (Source: Photo collection of Mrs A Mans)
mainly involved in Computer Science
lecturing. Mr Johan van Loggerenberg
was appointed in about 1973, also
as Scientific Programmer. Before
this appointment, he performed
programming and operating work for
the Department of Physics22.
In 1973, Ms Engela Wilson was
appointed as Punch Operator. Mr (later
Prof) T (Tjaart) Steyn started working at
the Computer Centre in January 1974.
In the 1970s, he played an important
role at the Computer Centre. Later
on, he moved to the Department of
Computer Science when it separated
from the Computer Centre.
Until 1970, Prof De Beer was only
involved with the Computer Centre
in a part-time capacity, since he still
had obligations at the Department of
Physics. From 1970 onwards, he was
transferred to the Computer Centre in
full-time capacity. Prof WP Robbertse,
the Dean of Natural Sciences,
advocated for the implementation
of Computer Science as a subject in
the Faculty of Natural Sciences,15 and
from 1971 Computer Science was
offered as a two-year major subject in
the Faculty of Natural Sciences.16 Prof
de Beer headed both departments
until 1978. While the two departments
functioned as a unit, the staff members
were allocated to one of the two
departments.
DEVELOPMENT OF
ADMINISTRATIVE
SYSTEMS
The most successful early
administrative application that was
developed for the University, was
the salary system that Ms Klaasje
Benadie developed in Fortran in
1968/69. The system was so
successful that it functioned until
1996, with necessary upgrades
and adjustments for the later
computer platforms performed as
required17. On the IBM 1130, the
salary system was mainly operated
using punched cards. Staff
members’ personal information,
their salaries and bonuses were
kept on punched cards, which was
updated, as required, by Mrs Bettie
Venter. Approximately 300 staff
members’ salary payments were
initially handled by the system.
This number increased to about
1600 in 1996.
Later on, a system was developed
to store and extract student
alumni’s contact information18.
Another administrative system that
was developed for the University
was the student assessment
system, which was developed by
1969. In Die Wapad of 26 June
1970, a report appeared on the
15 Transcription of a conversation between the Author and Prof JF de Beer, 15 January 2013
16 Faculty Board of Natural Sciences, minutes, Art 26, 18 February 1970 17 Transcription of the Author’s interview with Prof JF de Beer, Prof T Steyn and Mrs K
Benadie, 19 November 2008 18 ITB News, p 3, July 1992
14 15
statistical analysis that was done
on the 1969 exam results using the
computer19. From 1971 onwards, exam
results were printed and released using
the computer20.
The increased core memory of the
IBM 1130 allowed for the use of Cobol
as a programming language, which
enabled Mr Johan van Loggerenberg
to expand the student assessment
system into a basic student system21.
The extent of the programming for
administrative systems led to Mr
Johan van Loggerenberg and the
two Punch Operators being moved
to the Administrative Building where
they reported to the Registrar, Prof
HC van Rooy22. The Administrative
Development Group, that was based
in the Administrative Building until
1977 (when Mr van Loggerenberg left
the University’s employ), moved back
to Computer Services, from where
they operated from the beginning of
January 198022.
The Computer Centre also accepted
projects from the Potchefstroom
Municipality. In approximately 1973,
system, which was large for the time,
was successfully developed by Prof
De Beer and his team. This system
processed municipal bills for eleven
years and generated useful income
for the Computer Centre. In 1978, a
budget of R23 400 (approximately R700
000 in 2017 ZAR value) was forecast
for this work. This ongoing project
benefitted staff members by exposing
them to commercial systems and was
later on a very useful experience for the
Computer Science lecturers23.
IBM SOUTH AFRICA’S
PARTNERSHIP WITH
THE UNIVERSITY
IBM South Africa was involved with the
University’s Computer Centre in various
ways. In 1968, IBM presented a seminar
at the University24.
IBM South Africa made several
Figure 10 Mr JF Clarke, IBM South Africa CEO, discusses some IBM equipment with Prof JF de Beer during a seminar in 1968.
Figure 11 A donation from IBM South Africa is handed over to Prof JF de Beer (left) in 1972. (Source: Photo collection, NWU Archive)
the Computer Centre was approached
to process water and electricity bills
for the Potchefstroom residents. The
donations to the University in the
early years of the partnership. There
are photos of two such events in 1972.
Figure 12 Another donation is made by IBM South Africa to the University in 1972. In this photo, Mr HM Robinson, Head of Public Relations (second from the right) receives the donation in the presence of Prof JF de Beer (right). (Source: Photo collection, NWU archive.)
19 Die Wapad, Vol 25, No 9, p1, 2 June 1970 accessed at http://dspace. nwu.ac.za/bitstream/handle/10394/14348/Wapad_d25_n09_1970. pdf?sequence=1&isAllowed=y
20 Die Wapad, Vol 26, No 14, p6, 10 September 1971, accessed at http:// dspace.nwu.ac.za/bitstream/handle/10394/14440/Wapad_d26_n14_1971. pdf?sequence=1&isAllowed=y
21 Transcription of the Author’s conversation with Dr J van Loggerenberg, 14 October 2009
22 Transcription of the Author’s interview with Mr EH van Niekerk, 15 September 2009 23 Transcription of the Author’s interview with Prof JF de Beer, Prof T Steyn and Mrs K
Benadie, 19 November 2008 24 The photos in Figures 10, 11 and 12 with their captions obtained from the NWU
archive, is the only information available on the respective topics.
16 17
NEW LOCATION FOR
COMPUTER SERVICES
The Computer Centre and computers
were moved from the old location
in the Physics building to a newly
designed, fit-for-purpose location in
the Mathematics wing of the Natural
Sciences building at the beginning of
1974.
The two IBM 1130 computers were
in use until August 1974, when they
were replaced with an IBM 370/125
computer. Although the IBM 1130s had
limited calculation power and memory,
a lot of useful work was done on them.
The computer users had to learn how
to programme effectively, which is an
important skill - even when working
with modern computers with large
memories and powerful processors.
Figure 13 The wing of the Natural Sciences building where the Computer Centre was housed on the ground floor from 1974. The Department of Mathematics and Applied Mathematics occupied the top two floors.
CHAPTER 3 The IBM 370/125 computer, from 1974 to 1980
THE PURCHASE OF
THE IBM 370/125
On 18 September 1973, the Finance
Committee of the University Board,
on recommendation of the Computer
Committee, decided to obtain an IBM
370/115 computer at an annual rental
fee of R84 0001 (approximately R4,3
million per year in 2017 ZAR value).
At the time, the IBM 1130 had already
been in service for more than five years
(since 1967). This decision was upheld
by the University Council on 1973-09-
19 and an IBM 370/125 computer was
ordered.
The IBM 370/125 was installed during
August 1974. It had a memory of 128
kilobytes and removable disk drives
with a capacity of 75 megabytes
each. When it was found that the
operating system used almost all
the memory, it was upgraded to 256
kilobytes2. While the IBM 370/125 was
a substantial improvement on the
IBM 1130, it still could not handle the
demand for computer capacity and
interactive programme development.
1 Finance Committee, Minutes, 18 September 1973, Art 16. 2 Conversation between the Author and Prof T Steyn, 8 March 2017
19
20 21
This problem was highlighted in a
memorandum dated 5 April 19763,
by Dr (later Prof) H Moraal from
the Department of Physics, who
emphasised the importance of the
research method of using numerical
techniques in the natural sciences. The
use of these techniques created a lot of
opportunities for researchers, but had
high demands in terms of calculation
ability. It also required interactive
programming, which the IBM 370/125
was incapable of.
THE UNISA INVESTIGATION
In 1976, the University requested that
the Institute for Computer Systems
from the University of South Africa
Figure 1 Mr AJ (Johan) Smith
(UNISA) investigate the PU for CHE’s
computer infrastructure and make
recommendations4. (It is interesting to
note that four of the six members of
this task team were former students
of the PU for CHE’s Department of
Physics, namely Prof CH Bornman,
Prof N Coetzee, Prof NF du Plooy and
Prof JP Maree.) This institute presented
their report on 1 November 19765 and
made thorough recommendations
on the necessary business systems,
Computer Centre staff structure and
physical computer needs for the next
ten years. It was found that the demand
for computer capacity in 1976 was 1.8
times the existing capacity. The report
interestingly recommended that there
be a reasonable extent of integration
between the Computer Centre and the
Department of Computer Science, with
specialists being appointed jointly in
both departments. Shortly thereafter,
this was no longer applicable to the PU
for CHE, because the two departments
were separated in 1978.
On 1 April 19786, Mr AJ Smith was
appointed Head of the Computer
Centre, while Prof JF de Beer continued
working in full-time capacity as Head of
the Department of Computer Science.
After the two departments were
separated, the staff complement for
each department was as follows7:
Computer Science: JF de Beer, P
Bruwer, A Meyer, P Kotze and A Mans
Computer Centre: AJ Smith, K Benadie,
M Dent, T Steyn, J van Rooy, C Blume,
E Venter and E Wilson
Other recommendations made in the
UNISA report were partially
implemented:
• The recommendations regarding
the governance of the Computer
Centre were implemented. At
the end of 1977, the decision was
made to appoint a new Steering
Committee for the Computer
Centre with a sub-committee for
Administrative Systems and Library
and a sub-committee for Scientific
Users8. These committees were
appointed at the beginning of
19789. The Steering Committee
was constituted during its first
meeting on 10 April 1978. The Vice
Rector, Prof JS du Plessis, was the
Chairperson of the Committee,
which consisted of the Head of the
University’s Vaal Triangle Campus,
the two registrars, the heads of
the Computer Centre and the
Department of Computer Science,
as well as the chairpersons of the
sub-committees for Administration
and Library, and of Scientific Users.
• The IBM 370/125 was upgraded
at the end of 1977 with a central
processor that was 30% faster
and the memory was increased
to 512 kilobytes. A remote job
entry terminal was installed at the
Administration Building and five
computer terminals were supplied
to the Computer Centre10.
• A Data General Eclipse S230
minicomputer and several
computer term ina ls were
purchased to address the demand
for interact ive programme
development and for increased
processing speed for scientific
users and Computer Science
students. This computer was
delivered11 at the beginning of
1978 at a cost of R185 00012,
approximately R4,8 million in 2017
ZAR value.
• An IBM Series/1 minicomputer
was installed at the Vaal Triangle
Campus at the beginning of 197913.
It was connected to the mainframe
computer at the Potchefstroom
campus by means of a telephone
data line.
3 H Moraal, Memorandum regarding computer facilities, 5 April 1976 4 Policy and Development Council, Minutes, 1976-09-14, Art 3.7 5 University of South Africa, Institute for Computer Systems, Report regarding
computer utilisation at the PU for CHE, 1 November 1976 6 Computer Users Meeting, Minutes, 24 April 1978, Art 1 7 Giel Hattingh, Rekenaarwetenskap en Inligtingstelsels, Chapter 4.6 in ’n Bekroonde
Nalatenskap: Die Fakulteit Natuurwetenskappe oor 100 jaar, Editor: Elize S van Eeden, 2017
8 Computer Users Meeting, Minutes, Art 3.1, 9 November 1977 9 Letter from the Vice Rector addressed to the Author regarding his nomination as a
member of the Committee for Scientific Computer Users, 20 February 1978 10 Computer Users Meeting, Minutes, Art 3.1, 9 November 1977 11 Steering Committee, Computer Services, Minutes, Art 3, 25 January 1978 12 Computer Centre, Report, Art 2, July 1978, Appendix to Minutes, Steering
Committee, Computer Services, 7 August 1978 13 Computer Users Meeting, Minutes, Art 5, 31 January 1979
22 23
Figure 2 Prof JS du Plessis, Vice Rector of the University and Chairperson of the Steering Committee (right) receives the Data General Eclipse S230 minicomputer. (Source: Photo collection, NWU archive)
At the time of the UNISA investigation
in 1976, the staff complement was as
follows: The Group for Administrative
Systems separated from the rest and
Mr JJ van Loggerenberg and three
card punch operators reported to the
Academic Registrar and were located
in the Administrative Building. The
rest of the staff reported to Prof JF de
Beer and consisted of (i) the Group
for Scientific Systems and the Library
with Mr M Dent, Mr J Leonard and one
card punch operator, (ii) the Group
for Remuneration and Departmental
Development, which consisted of
Mrs K Benadie and (iii) the Group for
Computer Operation and External
Work that was manned by Mr T Steyn
and a computer operator. Thus, apart
from the two managers, there were five
computer specialists, four card punch
operators and one computer operator.
A second computer operator was
appointed at the beginning of 1978,
which allowed the daily operating
service time to be extended to 06:00
– 20:1514.
THE ECLIPSE MINI
COMPUTER
As mentioned earlier, the Eclipse S/230
minicomputer was purchased at the
beginning of 1978. While it was very
useful for calculation-intensive tasks, it
could only service a limited number of
users simultaneously. The Eclipse was
updated early in 1980 with an extra
hard drive and its core memory was
increased from 192 to 320 kilobytes to
enable it to service more users. By the
middle of 1980, terminals for the Eclipse
were also placed at the Departments
of Physics and Statistics15. Later in the
year, this facility was expanded and it
was possible to communicate with the
Eclipse from any location on campus
through acoustic modems16. The
limited number of simultaneous users
the Eclipse could serve did, however,
remain a problem. In 1981, the Eclipse
was no longer available for general
use and was reserved for experimental
work at Computer Science17.
In 1983, the Eclipse was donated to
the CSIR18. At this event, the Steering
Committee of Computer Services
decided to award a bonus of R1 000
to Andries de Beer, a computer science
student who did his best to optimally
manage and maintain the Eclipse 19.
14 Computer Users Report, Minutes, Art 7, 22 February 1978 15 Computer Services, Newsletter, August 1980 16 Computer Services, Newsletter, October 1980
17 Steering Committee Computer Services, Minutes, Art 5.3, 14 August 1981 18 Steering Committee, Computer Services, Minutes, Art 16.1, 27 June 1983 19 Steering Committee, Computer Services, Minutes, Art 17.1, 27 June 1983
24 25
THE MEDIA CONVERTER
After upgrading the IBM 370/125, there
were still problems with CPU overload.
One of the culprits was the load placed
on the central processor by relatively
slow peripherals, like the graphic
plotter, paper tape reader and other
equipment. In 1976, Electronic Services
started to develop small independent
microcomputer systems with Intel 8080
and 8085 microprocessors20 for use in
sophisticated research equipment.
Electronic Services was commissioned
to develop a microcomputer system
which could serve as media converter
for the IBM 370/125. This would serve
as an interface between the slow
peripherals and an IBM-compatible
digital magnetic tape recorder. Mr EP
Blignaut developed and programmed
this microcomputer system that was
commissioned at the end of January
197821. The IBM 370/125’s central
processor wrote the data for the
graphic plotter and other peripherals to
the (relatively fast) IBM magnetic tape
recorder. That eased the burden on the
IBM’s central processor by allocating
the slow mechanical plotting process
to the media converter and freeing the
IBM’s CPU from the slow peripherals.
Vice versa, the data from the punched
tape reader was transferred to
magnetic tape, which could be read
at high speed to the central processor
of the IBM 370/125 by one of the
IBM digital tape drives. The system
was expanded to enable the transfer
of data from the data cassette tapes
that, e.g. the Library used to capture
the book catalogue data to an IBM-
compatible digital tape recorder.
Figure 3 The media converter. On the photo, Mrs Renée Smit from the Department of Physics is transferring data from the punched paper tape to the IBM-compatible tape machine. The graph plotter is not visible on the photo. (Source: Photo collection, Mrs AD Mans)
THE CAFETERIA SYSTEM
In the 1970s, a facility was developed
for computer users where they could,
for the first time, read punch cards into
the computer to run small computer
tasks. This facility included card punch
machines, a card reader and a printer.
It is not clear when this facility was
established, but it was first mentioned
in the minutes of the Computer Users
Meeting on 28 September 197722
when it was announced that a third
card punch would be installed in
the cafeteria. Through the cafeteria,
Computer Science students and
researchers could punch and test
short programmes by themselves. The
Author used this system and found it
very useful for short tasks, unlike the
batch processing system that required
that all computer tasks be qued at Task
Reception.
COMPUTER USERS
By mid-1978, there were approximately
200 computer users at the University.
The Committee for Scientific Users
remarked that this number was much
lower than that of the University of
the Witwatersrand23. According to the
Computer Centre’s report dated 6 June
1978, scientific calculation took up 42%
of computer time and administrative
applications 10%. At that stage, there
were six specialists and four operators
on the staff24.
The Computer Users Meetings started
on 8 June 1977 and meetings were
held monthly. These meetings were
useful networking opportunities for
Computer Centre staff and computer
users.
The last minutes that could be found of
such a meeting was dated 2 June 1981.
It is not known whether these meetings
still took place thereafter.
20 Committee for Scientific Users, Minutes, 19 November 1979, Appendix: ʼn Samevatting van die huidige verspreiding en standaard van plaaslik ontwikkelde mikroverwerkingstelsels, compiled by Electronic Services, 19 November 1979
21 Computer Users Meeting, Minutes, Art 8, 25 January 1978
22 Computer Users Meeting, Minutes, Art 4, 8 June 1977 23 Committee for Scientific Users, Minutes, Appendix: Document detailing the
planning of the Computer Centre from September 1978 to September 1979, Art IV-3, 23 August 1978
24 Steering Committee, Computer Services, Minutes, Appendix, 6 June 1978
26 27
THE COMPUTER
CENTRE BECOMES
COMPUTER SERVICES
By mid-1979, the name of the
Computer Centre was changed to the
Department of Computer Services
and Mr AJ Smith’s job title changed to
Director of Computer Services. At this
stage, the department’s most popular
service was the processing and marking
of multiple-choice question papers25.
In time the IBM 370/125’s age and
annual maintenance costs increased
significantly. The bigger problem,
however, was its inability to meet
the demand for computing power.
Computer Services had a very limited
budget in the late 1970s, making it
impossible to buy a new computer.
The fact that all programmes and data
had to be uploaded to the computer
with punched cards, remained a
problem. Smaller programmes could,
as mentioned earlier, be uploaded by
users at the cafeteria. Big tasks had
to, however, be submitted on punch
cards at Task Reception, where there
was often a waiting period.
Prof Tjaart Steyn preserved some
historical computer components from
the earlier years, including the black
board depicted in Figure 5 that was
used to indicate the waiting time
for the different task streams. The
task stream for the cafeteria had the
highest priority, but was available
only for very short programmes. BG
was the next priority and was used
for programmes requiring up to 10
minutes for calculation. F3 was used
for longer programmes26.
Figure 4 The management of the University visited the machine room in April 1979. From left to right, Prof T van der Walt (Rector), Prof JS du Plessis (Vice Rector), Prof WP Robbertse (Dean of Natural Science), Mr J Smith (Director of Computer Services) and Prof JF de Beer (Head, Department of Computer Science). (Source: Photo collection, NWU archive)
Figure 5 The blackboard that indicated the expected waiting times for the different task streams during the IBM 370/125 era. As can be seen, a user indicated frustration by exaggerating the values.
25 Computer Services, Newsletter, Vol 2, Nr 3, November 1979 26 Computer Centre, Report, Art 2, 6 June 1978, Appendix to the Minutes of the Computer Services Steering Committee, 6 June 1978
28 29
Figure 6 The west wing of the computer room in May 1979. The disk drives are visible on the left. Mrs Anne Mans is in the background and Mrs Klaasje Benadie stands with her back to the camera. (Source: Photo collection, NWU archive)
Figure 7 The east wing of the computer room in May 1979. Mrs Anne Mans is standing by the punched card reader. (Source: Photo collection, NWU archive)
NEWSLETTER
In the 1970s, a newsletter for computer
users was published for the first time. It
is unknown when the first edition was
published, but the second edition was
published in 1978. This newsletter was
published several times a year. The first
few editions consisted of a single page,
while later editions consisted of several
A4 pages that were stapled together.
From 1979, the newsletter had its own
front cover, which was later a stylised
PU for CHE emblem printed with a line
printer (see Figure 8).
The Computer Services newsletter
contained valuable information, as
well as new developments in
computer science. Later on, it also
contained news from other universities’
newsletters. The last available edition
of this newsletter is the August 1980
edition. Although an edition was
prepared for October 1980, it was,
according to a note attached thereto,
never published. The newsletter was
succeeded in 1984 by RKD Nuus that
was published and bound in A5 format.
Figure 8 Example of Computer Services’ newsletter front page in November 1979.
30 31
OPEN DAYS
On 27 June 197927 and 10 September
198028 Computer Services hosted
open days to present modern
technology and developments to
users. At the second such open day,
the following developments were
presented:
• Sm a l l H e w l e t t - P a c k a r d
calculators, like the HP-35, HP-
45, HP-65, etc.
• Apple II home computer, which
was released in South Africa at
Figure 9 In this photo from the June 1979 open day, equipment is being demonstrated to Prof J Spoelstra from the Department of Mathematics (sitting) and Prof PH Stoker from the Department of Physics (middle back).
about that time
• DEC, ICL and IBM commercial
computers
• IBM and CDS word processors
• Tektronix graphics screens
• Electronic Services showcased
some of their locally-developed
microcomputers
The Computer Services Meeting
considered the open days very
successful29.
FURTHER DEVELOPMENTS
At the beginning of 1980, the highly-
regarded mathematical/statistical set
of IMSL subroutines were purchased for
mathematical and statistical analyses.
The WATFIV Fortran translator was
also made available and it streamlined
programme development and
corrections30. Further expansion
included a graphic plotter with a 36-
inch drum by the end of 198031 and a
digitiser, which had been purchased in
197932.
At the Vaal River Branch (later the
Vaal Triangle Campus), a demand
developed for computer facilities,
especially after the appointment of Prof
JM Hatting who lectured Operational
Research, Mathematics and Computer
Science at that campus of the University.
Initially, in 1978, a computer terminal
was provided, but in the beginning of
1979 an IBM Series/1 minicomputer33
was provided to the Goodyear Street
Campus and was connected to the IBM
370/125 with a post office data line.
This computer was especially used by
the Computer Science students.
A lack of space was a continuous
problem for Computer Services and
was often discussed by the Computer
Services Steering Committee. At the
Committee’s meeting on 22 November
1979, Prof WP Robbertse mentioned
that the space used for the old nuclear
accelerator could possibly be used for
Computer Services34, something which
was only realised ten years later to
solve a very pressing space problem.
Mr AJ Smith, Director of Computer
Services, left the employ of the
University at the end of November
198035. He was the first director
who had previously worked in the
professional computer environment.
Further expansion of Computer
Services during his term was prevented
by an extremely tight budget. He was
succeeded by Prof JM (Giel) Hattingh,
who was appointed at the beginning
of 198136.
The 1970s were trying years for
Computer Services with the IBM
370/125 that had limited computing
power and could not address the
growing demands of the computer
users. The IBM 370/125 was in service
from August 1974 to July 1981; a long
time to struggle with a computer with
inadequate computing power. Prof
Hatting prioritised addressing this
problem in the 1980s.
27 Computer Users Meeting, Minutes, Art 5, 26 July 1979 28 Computer Services, Newsletter, August 1980 29 Computer Users Meeting, Minutes, Art 6, 1 August 1979 30 Computer Users Meeting, Minutes, Art 4a, 5 March 1980 31 Computer Users Meeting, Minutes, Appendix 1, 8 October 1980
32 Computer Services, Newsletter, November 1979 33 Computer Users Meeting, Minutes, Art 5, 31 January 1979 34 Steering Committee, Computer Services, Minutes, Art 4.2c, 22 November 1979 35 Letter from Mr AJ Smith to the Author regarding farewell event, 10 November 1980 36 Computer Users Meeting, Minutes, Art 4, 3 March 1981
4
CHAPTER
The 1980s (1981-1990)
INTRODUCTION
During the 1980s, the mainframe
computer environment expanded
quite significantly. Batch processing
with punched cards as the general
input source for external data was
slowly phased out. The new approach
would be to use interactive computer
terminals that could be distributed
across the University’s campuses.
This was also the beginning of inter-
university networks, with connections
being established between the
mainframe computers of all South
African universities. This would later on
be expanded to include international
universities.
With the expansion of interactive
computer facilities, computer use
became common in many academic and
administrative departments, especially
as a result of real-time administrative
systems, the more widespread use
of scientific calculations and the
increasing use of word processing on
the mainframe. This further increased
the demand for computing power, and
this necessitated frequent upgrading
of the mainframe computers. This had
to happen despite budget pressure
caused by the strong decline in the
value of the South African Rand (ZAR) in
the 1980s that caused severe pressure
33
34 35
on university budgets. Between 1981
and 1985, the ZAR value declined by
approximately 150%, with the biggest
decline taking place between 1984 and
1985 (see Figure 1). Most computer
equipment was imported from the
USA and any decline in ZAR value had a
proportional impact on the University’s
dollar-based computer capital and
maintenance costs. The 1980s was
thus a time of restricted budgets and a
growing demand for faster computers
At the end of the 1980s, there was a
strong drive to develop local area
networks (LANs) on the two campuses.
This would later lighten the load on the
expensive mainframe environment by
transferring the computing load to
personal computers (PCs) and servers
on the LANs.
An important development in this
decade was the rise of microcomputers,
from a humble beginning to a
dominating movement that would later
replace the mainframe computer almost
entirely. The development of LANs,
which connected microcomputers,
servers, powerful departmental
computers and mainframe computers
with one another, ushered in the
1990s. Sometime later, the mainframe
environment became redundant at
most universities.
At the end of 1990, the University’s
computer-related departments
were reorganised to develop an
a l l -encompassing department
of Information Technology and
Management (ITM). This chapter
describes the history from 1981 to
just before the formation of the new
department of Information Technology
and Management.
Figure 1 The ZAR-USD exchange rate in the 1980s1. The significant decline in ZAR value from 1981 is evident and had a significant impact on the finances of Computer Services, since the bulk of their capital and equipment maintenance costs was US Dollar-based.
GOVERNANCE,
MANAGEMENT AND STAFF
Council and Management
Committee
The Management Committee of the
University had to approve all important
decisions regarding computer services.
All decisions with a significant financial
or organisational impact had also to be
approved by the University Council.
Empowering the Management
Committee by providing information
on computer science, was important.
IBM2. for example, presented an
overview of computer technology to
the Management Committee on 5 and
6 May 1981.
Steering Committee
The Director of Computer Services
reported to the chairperson of the
Steering Committee of Computer
Services, the Vice Rector of Finance
and Management.
It is unclear when the first Steering
Committee for Computer Services
w a s e s ta bl i s h e d, bu t o n 1 8
September 1973, during a meeting
of the Committee on Finance of the
University Board, mention was made
of the Computer Committee. At that
time there was, therefore, already a
type of committee that advised the
Computer Centre. It also appears that,
even before obtaining the first IBM
1130, there was already a committee
that had managed the procurement
process. As mentioned in a previous
chapter, the UNISA investigation team
of 1976 recommended that a Steering
Committee for Computer Services be
instated with relevant subcommittees.
The first documented minutes of
the Computer Services Steering
Committee that could be obtained was
that of 10 April 1978, which was chaired
by the Vice Rector, Prof JS du Plessis.
During the meeting, it was mentioned
that the committee was established
by the Rector on the grounds of an
unknown memorandum addressed
to him. Two subcommittees, namely
a committee for Scientific Users and
for Administrative and Library systems
were also established. At the following
1 Numerical values obtained from https://businesstech.co.za/news/finance/116372/ rand-vs-the-dollar-1978-2016/
2 ISM – n Terugblik, deur P du B Weʼnman Beheerkomitee Rekenaardienste, Appendix to Minutes, Art 6, 5 July 1984
36 37
meeting of the Executive Committee
of the Steering Committee, the Head
of the Computer Centre, Mr AJ Smith,
was instructed to compile a set of
rules and regulations3. for Computer
Services. His proposed rules and
regulations were approved by the
Steering Committee of Computer
Services on 3 September 19794. The
University Council approved the rules
and regulations on 13 September 19795,
which appeared as an Appendix to the
minutes of the Steering Committee of
Computer Services meeting held on 22
November 1979.
The rules and regulations, as well as
the composition of the Steering
Committee of Computer Services
were later amended, amongst others,
in 19816.
In 1984, a new Steering Committee
was appointed, and new rules and
regulations were accepted7. Provision
was made for two subcommittees,
namely Research and Education.
In the late 1980s, the Steering
Committee included a large group of
senior staff members. From the last
available minutes of the Committee (7
November 1988), it seems as if the
Committee consisted of 29 members.
The Steering Committee of Computer
Services was an important forum, where
the planning, budget and functioning
of Computer Services were discussed
at a high level by interested parties and
specialists.
Management
In the previous chapters, it was indicated
that Prof JF (Hannes) de Beer was the
first Head of the Computer Centre
which, together with the Department
of Computer Science, functioned as a
unit. In 1978, the two departments were
separated and Mr AJ (Johan) Smith was
appointed as Head of the Computer
Centre. The name “Computer Centre”
was changed to Computer Services in
1979 and Mr Smith’s title was changed
to Director. Mr Smith left the employ of
the University at the end of November
1980.
At the beginning of 1981, Prof JM (Giel)
Hattingh was appointed as Director
of Computer Services. Prof Hattingh
brought the problem of limited
computing power to the attention of
University Management. He ascribed
it to the fact that Computer Services
was financially neglected in the 1970s8.
From the Vice Rector of Finance and
Management’s communication to Prof
Hattingh (at that time the Director
of Finance and Planning) dated 20
October 1986, it appears that Prof
Hattingh’s continuous representations
were successful, as it is noted that the
University allocated double the amount
provided for in the subsidy formula to
Computer Services in 19869 (although
it still was not nearly enough to ensure
a quick response on the mainframe
computer). Prof Hattingh is known for
the role he played in the expansion of
the mainframe computer environment.
In 1985, he was promoted to Director of
Finances and Planning, but remained
involved in computer matters. From 24
June 1986, he was the Chairperson for
the Steering Committee of Computer
Services 10 .
Figure 2 Prof JM (Giel) Hattingh
Prof Hattingh was succeeded as
Director by Dr P du B (Philip) Welman.
Dr Welman was familiar with the data
network environment and played a key
role in establishing local area networks
at the University and in developing a
national academic data network in
South Africa. He also championed the
process of diversifying the University’s
computer p lat forms, adding
departmental and microcomputers.
He was in the University’s employ until
October 1990.
Figure 3 Dr P du B (Philip) Welman, Prof JM Hattingh’s successor as Director of Computer Services
3 Steering Committee of Computer ServicesSteering, Minutes, Art 4.1, 8 September 1978
4 Steering Committee of Computer Services, Minutes, Art 4.1, 3 July 1979 5 Steering Committee of Computer Services, Minutes, Art 4.1, 22 November 1979 6
Steering Committee of Computer Services, Minutes, Art 2, 14 August 1981
7 Steering Committee of Computer Services, Minutes, Art 7.3, 13 February 1984 8 Computer Services, Strategic Plan, Art 2, August 1982 9 Letter, NJ Swart to JM Hattingh, 20 October 1986 10 Executive of Steering Committee of Computer Services, Minutes, 24 June 1986
38 39
Computer Services
sub-departments
In the early 1980s, Computer Services
consisted of two divisions, namely
the divison of Academic Systems and
that of Administrative Systems, each
of which had a Deputy Director in
charge11.In 1984, these divisions were
reorganised and another division was
added. Thereafter the three divisions
were Facilities, Support Service and
Administrative Information Systems.
This structure was maintained until 1988
when the division of Administrative
Information Systems was moved from
Computer Services to the Department
of Management and Information
Services. For the remainder of the
1980s, Computer Services consisted of
only the Facilities and Support Services
departments.
Figure 4 Mr Gabriel Makhonxa (left) conducting operating work in the machine room at Computer Services. (Source: Photo collection of Mrs AD Mans)
Staff
By mid-1981, the staff complement of
Computer Services increased to 25,
10 of which were specialists. It was,
however, far less than Universities
like the University of the Orange
Free State (UOFS), which had 41 staff
members in their computer services
department. This was brought to the
attention of the Vice Rector of Finance
and Management by Prof Hattingh
in a letter dated 27 July 198112.
The staff shortage and insufficient
remuneration for computer staff
remained a problem at Computer
Services for a long time. This resulted
in the delay of the redevelopment of
interactive administrative systems,
which were intended to replace the
batch processing systems.
In the Computer Services newsletter,
RDK Nuus, which was published
regularly from 1984 onwards, staff
members were listed, making it easier
to keep track of the staff complement
growth. In 1984, the number of
positions increased to 38, 21 of whom
were specialists13. In 1985, there were
41 positions, with 23 specialists; in
1986 there were, respectively, 42 and
23 positions; in 1987, these numbers
were 40 and 23; and in 1988, this
number changed to 26 and 1 after the
division of Administrative Information
Services moved to the Department of
Management and Information Services.
In 1989, the Department of Computer
Services had 25 staff members, 18 of
whom were specialists.
In the late 1980s, an interesting staff
matter presented itself in the form of Dr
Philip Welman who went out of his way
to create a position for Mr GB (Gabriel)
Makhonxa at Computer Services.
Initially, Mr Makhonxa worked as a
messenger at the Bureau of University
Education. Several staff members,
including Prof Peter van Eldik (the
Dean of Natural Sciences at that time)
and Dr Welman noticed Mr Makhonxa’s
aptitude. Prof van Eldik helped him to
complete Grade 12 by mentoring him
in Mathematics, and Dr Welman
appointed him as contact person at
Task Reception on 1 February 1996. It
became clear quite early on that he did
very well in this working environment
and was trained and moved to the
Operations Division where he worked
as computer operator. Dr Welman
had a lot of difficulty appointing him
as a technical staff member, since he
originally worked as a messenger. With
the help of the Rector at the time, Prof
CJ Reinecke, the appointment was
finally made14. (Mr Makhonxa was the
first black person who was appointed
as a regular non-academic staff
member and not a service worker at
the University.) Mr Makhonxa continued
with his studies and obtained a BCom
Degree and a Diploma in Higher
Education. Although he had a bright
future in the computer world, he chose
to become a teacher, as he felt this was
the best way to help his people. He left
the employ of the University at the end
of 199215.
There are not many photos available of
the staff members from the 1980s. The
following are some of the rare photos
available:
11 Steering Committee of Computer Services, Minutes, Appendix regarding staff needs, 14 August 1981
12 Letter from JM Hattingh to NJ Swart, 27 July 1981
13 RKD Nuus, p 9 and 10, November 1984 14 Transcription of the Author’s interview with Dr P du B Welman on 15 October 200 15 ITM, Annual Report, Art 2.4.4, 1992
40 41
Figure 5 Mrs Elsie Buys was a central figure in Computer Services and ITM for a long time. She was the secretary for the Director of Computer Services
Figure 7 Mr CML (Krap) Steinmann was a well-known operator in the 1980s. (Source: Photo collection of Mrs AD Mans)
MAINFRAME COMPUTER
EQUIPMENT AND
SOFTWARE
The IBM 4331 computer
Prof JM Hattingh quickly made a strong
case for the purchase of a computer
that would be capable of supporting
interactive processing through
terminals. He indicated that the batch
processing method, which, at that
stage was done with the IBM 370/125,
was a very ineffective way to use staff
members in the User Departments
and in Computer Services. At the
Steering Committee of Computer
Services meeting of 6 March 1981, it
was recommended that an IBM
4331 computer be purchased16. This
which was much more powerful. It had
a 3-megabyte memory and 2 gigabytes
of disk space18.
One of the biggest advantages of the
IBM 4331 computer was its ability to
run the VM/SP operating system, which
meant that the computer could, with
the help of interactive terminals, be
operated from outside of the computer
room. This eliminated the hassle
of punch cards that were punched,
handed in at Task Reception, where
it had to be collected once the data
from approximately 1979 to July 1993. (Source: Photo collection of Mrs AD Mans)
recommendation was approved by the
University Council on 2 April 198117. In
July 1981, the IBM 370/125 was replaced
with the IBM 4331 mainframe computer
was processed and the results were available. Now, users could upload
programmes directly from a computer
terminal and see the results fairly soon,
Figure 6 Mrs Chrissie Blume, Mrs Anne Mans and Mrs Klaasje Benadie were all central figures in Computer Services and, later, ITM. (Source: Photo collection of Mrs AD Mans)
Figure 8 One of the early computer terminals that display the University emblem. ( S o u r c e : P h o t o collection of Mrs AD Mans)
16 Steering Committee of Computer Services, Minutes, Art 5, 6 March 1981 17 University Council, Minutes, Art 16, 2 April 1981 18 ITB Nuus, Memoires, p 4, July 1992
42 43
Figure 10 The photo of Computer Services staff members in the machine room, which appeared in IBM Inform. From left to right: Mrs Klaasje Benadie, Mr Johan Nel, Mr Jan van Rooy, Mrs Anne Mans and Mr Richard Drevin. (Photo, with acknowledgement, from IBM Inform, April/May 1983)
unless the programme was particularly
long. Interactive terminals were made
available to users from April 198219. A
total of 35 terminals were originally
purchased, four of which had colour
graphics screens. Initially, there were
20 terminals available to users.
Further computing facilities was made
available to scientific users when a
CDC terminal was connected to the
CSIR computer in July 1981 by means
of a data line20.
The VM/SP operating systems enabled
the University’s administrative systems
that, at the time, had to rely on
batch processing, to be converted to
interactive systems. This resulted in
the large-scale redevelopment of the
administrative systems, which put a lot
of pressure on the development staff
of Computer Services.
The interactive use of the computer by
many users put the central processor
under a lot of stress. As such, the
processor was upgraded to an IBM
4331-K2, which increased processing
speed by approximately 80%. This
resulted in several improvements and
allowed Computer Services to start
planning interactive administrative
systems.
The University’s computer set-up at
the beginning of 1983 is described in
detail in an article that was published in
19 Computer Services, Report, Junie 1982 20 Computer Users Meeting, Report, 2 June 1981
44 45
Figure 9 Mrs Anne Mans at the workstation that was connected to the CSIR’s powerful computer. To her left is the punch card reader. This station also had a printer that is not visible on this photo. (Source: Photo collection of Mrs AD Mans)
IBM Inform’s April/May edition21. (IBM
Inform was IBM South Africa’s magazine
that reported on new developments
and gave details of clients’ computer
installations.) According to this article,
the University’s IBM 4331 computer had
a 3-megabyte memory and 2 gigabytes
of disk space.
The use of terminals by end users
resulted in the availability of
applications like word processing,
stimulated more general use of the
computer, also by clerical staff.
In April 1983, the Focus database and
development software for business
systems was purchased. In July of
the same year, the development
of an interactive student system in
Focus commenced. This added more
stress to the computer, but enabled
the elimination of the clumsy batch
processing systems.
One of the most important facilities of
the Computer Centre was Task
Reception, especially when batch
processing was still commonly used
for big tasks and printers were not
widely available. The photo in Figure
11 depicts Task Reception in the early
1980s. Mrs Elise van Dijken worked
there for quite some time.
Figure 12 The IBM 1403 printer was a vital component at Task Reception. It was a fast line printer that could print approximately 1 000 lines per minute. (Source: Photo collection of Mrs AD Mans)
Figure 11 Task Reception, with Mrs Elise van Dijken on the left of the photo. (Source: Photo collection of Mrs AD Mans)
The IBM 4341
The demand for computing power
continued to increase and in July 1983,
the University upgraded to an IBM 4341
Model 10 with a 4-megabyte memory,
which, in turn, was upgraded to an IBM
4341 Model L2 with an 8-megabyte
memory in January 1984. This IBM
4341 had an IBM 3080 disk controller
and an IBM 3705 communication
controller, which lessened the load of
peripheral equipment on the CPU
quite considerably.
The highly recommended and extensive
set of subroutines for mathematical
and statistical work, IMSL, was installed
in 198422. This addressed a big need in
scientific programming and remained a
very popular resource. Another useful
programme that was released during
this time was IBM’s Professional Office
System (PROFS), which was used for
office automation23.
A notable improvement was the
installation of two IBM 6670 laser
printers in 1983, one at Computer
Services and one in the Administration
building. These were the first high-
volume laser printers at the University,
which had the added benefit of
being able to print Afrikaans text
with diacritical characters, which was
imperative in letter writing in Afrikaans.
21 IBM Inform, “PU hou tred met rekenaartegnologie”, April/May 1983, p14-15 22 RKD Nuus, p16, November 1984 23 RKD Nuus, p11, November 1984
46 47
The IBM 4381 and later
computer platform
diversification
The number of mainframe computer
users continued to increase and in
August 1984 there were 800 active
users, which again created response-
time problems. As a result, the IBM
4341 was replaced with an IBM 4381,
which was much more powerful24. This
computer’s physical size required that
the computer room be extended,
which, in turn, required that the two
adjacent offices be incorporated into
the computer room. The walls to these
offices were broken down and a false
floor was installed.
The IBM 4381, with its accompanying
tape and disk drives, was commissioned
on 22 January 1985. This was a long,
complicated process that started on
7 January 1985 and necessitated the
total reorganisation and expansion
of the computer room25. The new
computer had a 16-megabyte memory
(double that of the IBM 4341) and the
processing speed more than doubled.
After the new computer was installed,
IBM South Africa hosted a social
reception for the staff members of
Computer Services at the University’s
farm, Nooitgedacht. IBM South Africa’s
Berthol Theiland, who served as IBM’s
sales representative at the University
for quite some time, played a key role
in the computer installation and social
reception.
Figure 14 Mr Berthol Theiland served as a sales representative of IBM South Africa at the University for a long time. In this photo, he is working at the social reception that IBM South Africa hosted for the Computer Services staff after the IBM 4381 installation. (Source: Photo collection of Mrs AD Mans)
Figure 15 Computer Services staff members enjoying the reception at Nooitgedacht. (Source: Photo collection of Mrs AD Mans)
Figure 13 The computer room in April 1985. (Source: NWU archive)
Figure 16 Mrs Wilna Oosthuizen and Mrs Klaasje Benadie in a relaxed mood at the Nooitgedacht reception. (Source: Photo collection of Mrs AD Mans)
Figure 17 Mr Jan van Rooy at the Nooitgedacht reception. (Source: Photo collection of Mrs AD Mans)
24 Executive Meeting of Steering Committee of Computer Services, Minutes, 12 March
1986, in Steering Committee of Computer Services, Agenda, 21 October 1986
25 RKD Nuus, p1, March 1985
48 49
From the 1982 implementation of
the IBM 4300 series of mainframe
computers until 1985, there was a
dramatic increase in computing power
(see Figure 18).
Increasing computer power became
a necessity when the number of
computer terminals increased as a
result of an increase in user numbers.
Figure 19 depicts the increase in the
number of terminals between 1982 and
1985.
The popularity of the mainframe
computer continued to increase and
in August 1985, the number of active
users totalled 1 00026.
In August 1985, Computer Services
was visited by Prof Heinrich Herbstreith
from the University of Karlsruhe in
Germany. Prof Hattingh invited him
to advise Computer Services on the
planning of the University’s computer
infrastructure. His recommendations
were summarised in Dr Philip Welman’s
presentation, Strategy for the use of
computers, which was presented on
21 November 198627 at the Steering
Committee of Computer Services
meeting.
Prof Herbstreith recommended that
the University diversify its computer
set-up and, apart from the mainframe
environment, also provide for
Figure 18 The increase in processing power of the University’s IBM 4300 range of computers between 1982 and 1985 (Source: Computer Services, Annual Report, July 1984 to June 1985, p12, annexe to Management Committee of Computer Services, meeting minutes, 25 November 1985)
Figure 19 The increase in the number of computer terminals at the University between 1982 and 1985. (Source: Computer Services, Annual Report, July 1984 to June 1985, p15, annexe to the Management Committee of Computer Services, meeting minutes, 25 October 1985)
departmental computers and for a
microcomputer on every desk. All
these systems had to be connected to
national and international networks.
Independent computers had to be
avoided entirely, and good training
equipment had to be prioritised
for Computer Science. These
recommendations were accepted by
the Steering Committee28 and had a
significant impact on the future
planning of the University’s computer
infrastructure.
In 1985 the first step was made in the
direction of data communication with
other universities when a connection
was made to the University of Pretoria’s
mainframe computer. This was very
useful for the Ferdinand Postma library
that could start exploring the Dobis/
Libis library system on the UP computer.
(See Chapter 6 for more information on
the library’s computerisation.)
In February 1986, the number of active
computer users increased to 1 200.
The continuously growing computer
use resulted in serious problems
regarding computer response time in
1986. This happened even though the
University upgraded to the IBM 4381
in 1985 and from the VM/SP operating
system to VM/HPO in April 1986. One
of the biggest reasons for this problem
was the real-time student system that
took up approximately one fifth of the
computer capacity29. The impact of the
long computer response time was so
significant, that it was discussed at the
University’s Management Committee
on more than one occasion30. The
University’s budget was, however,
under severe pressure and the Vice
Rector of Finance and Operations, Prof
NJ Swart, wrote a letter to Prof Hattingh
on 20 October 1986, in which he
informed him that Computer Services
was allocated a budget of R2,5 million
for both 1987 and 198831. In addition,
the Steering Committee of Computer
Services decided in October 1986 that
the installation of a PC laboratory at
the Potchefstroom campus and one
at the Vaal Triangle Campus should
be prioritised and that the mainframe
computer upgrading had to wait32.
In future, more emphasis would be
placed on distributing the computing
load to other platforms, rather than
upgrading the mainframe computer.
This necessitated the implementation
of good data networks, which would, in
future, receive a lot of attention.
An important expansion in the
mainframe computer software was
the implementation of Graphics Data
Display Manager (GDDM), SAS in 1985
and Storage and Information Retrieval
System (STAIRS) in 198633. GDDM could
be used to display data graphically on
a computer terminal with graphics
26 RKD Nuus, p1, November 1985 27 Steering Committee of Computer Services, agenda, 21 October 1986 28 Steering Committee of Computer Services, Minutes, Art 5.4, 21 October 1986 29 Computer Services, Annual Report, Art 2, 1986
30 Steering Committee of Computer Services, Minutes, Art 16.2, 27 February 1986 and Art 9, 18 March 1986
31 A letter from NJ Swart to JM Hattingh, 20 October 1986 32 Steering Committee of Computer Services, Minutes, Art 5.9, 21 November 1986 33 RKD Nuus, Chapter 3.1, April 1986
50 51
capabilities. SAS provided extensive
support to, amongst others, statistical
analysis and extended graphic abilities.
STAIRS could store unformatted
information, like regular documents,
in a database. It had a powerful search
function.
In 1987 the Mohawk data capture
system 34 , that functioned with
diskettes instead of punched cards,
was implemented. The diskettes thus
replaced the punched cards as the
most important data input medium for
batch processing systems.
In mid-1987 it was reported that the
number of active computer users
declined by 200, which was ascribed
to the fact that the two PC laboratories
were taken into use the previous year
by Computer Science students who
were, previously, dependent on the
mainframe computer. The problem
with response time was, however,
not addressed and in June 1987, the
IBM 4381 P02 was upgraded to an
IBM 4381-Q14. This computer had
two processors and a 24-megabyte
memory. The total disk space was
now 7.5 gigabytes, and there were
345 terminals35 at the University. The
computer’s response time improved
significantly with the upgrade and the
real-time student registration in 1988,
which included the issue of receipts for
the first time36, was completed without
incident. For this upgrade, the machine
room had to be increased in size by
taking up two more offices, which
exacerbated the office space problem
for Computer Services. Some relief
came from the two new computer labs
for students that were installed in the
accelerator building and that served
as microcomputer lab, graphics lab
and terminal lab on the first floor and
a training room and demonstration
centre on the second floor37.
In 1988, a lot of damage was caused
by a fire in the student terminal room,
possibly as a result of a cigarette butt
that was disposed of in a scrap paper
container38. The damage was
estimated at R100 000 for equipment
and approximately R90 000 for the
building. A total of 24 terminals and a
printer was destroyed.
In April 1988, Afrikaans document
proofreading was introduced39. This
could be done on the mainframe
computer by using a combination of
Script and Displaywrite/370. Interesting
to note is that the work on this started
when IBM South Africa commissioned
Prof Hertzog Venter, a retired Professor
in Afrikaans/Dutch at the PU for CHE,
to compile an Afrikaans wordlist with
hyphenation. He completed this
wordlist, but passed away on 17 May
1988 before he could complete the
synonym list40.
The stress on the mainframe computer
was further decreased with the
installation of two departmental
computers; one was installed in the last
term of 1988 and one in the beginning
of 1989. The first computer was an IBM
9375 Model 40 with a 16-megabyte
memory and 1.7 gigabyte disk space
for the Departments of Physics and
Statistics41. It was installed at the
Department of Physics and was
intended to function as a powerful
processor for the computer-intensive
research work conducted in the
aforementioned departments. This
computer did, however, not meet
expectations and was soon updated
to a model 6042, which more than
doubled the speed. The users needed
even more computing speed and
the IBM 9375 was replaced with an
RS/6000 RISC processor late in 1990
(more on this later in this chapter).
Figure 20 Damage caused in early 1988 by a fire in the student terminal room (Source: Photo collection of Mrs AD Mans)
Figure 21 The hand-over of the IBM 9375 computer to the Departments of Physics and Statistics. From left to right, Prof Harm Moraal, Prof Peter van Eldik, Dr Philip Welman, an ISM representative, Prof Giel Hattingh, another ISM representative and Prof Pieter Stoker. (Source: Photo collection, NWU archive)
34 RKD Nuus, p18, Julie 1987 35 RKD Nuus, p31, July 1987 36 RKD Nuus, p30, April 1988 37 RKD Nuus, p2-8, October 1987 38 RKD Nuus, p34, April 1988
39 RKD Nuus, p20, April 1988 40 Private communication with Ms Mariet Venter, daughter of prof Hertzog Venter, 5
May 2017 41 RKD Nuus, p10, March 1989 42 RKD Nuus, p6, July 1989
52 53
The second departmental computer
was an IBM 4361 Model 3 with a 4-
megabyte memory and 1 040
megabyte disk space. This computer
(donated by Volkskas Aksepbank43)
was installed at the Vaal Triangle
Campus in Vanderbijlpark. This large
F i g u r e 22 The installation of the IBM 4361 at the Vaal Triangle Campus. The computer
piece of equipment had to be hoisted
three storeys high and moved into the
building from the outside with a crane.
This installation was completed on 13
January 1989. The computer was mainly
used by the students of the Computer
Science and Engineering departments44.
In late 1988, an IBM 7171 protocol
converter was purchased46, making it
easier for users to work from home
and, therefore, during quieter times.
This converter provided access to the
mainframe computer through dial-
up modems. This allowed users to
access the mainframe from their home
PCs through a terminal emulation
programme and a suitable modem.
Initially, four dial-up modems were 47
the RKD Nuus of March 1990, André
de Waal from Computer Sciences
motivated users to use electronic mail,
because “it could change their lives”.
Prophetic words!
The space problem in the Computer
Services machine room, was finally
solved when the ground floor of the
old accelerator centre in the last half
of 1989 was allocated for use as the
machine and control room. This room had to be hoisted three storeys high and moved
provided . had a surface area of 400 m2 and could
into the building with a crane. (Source: Photo collection of Mrs AD Mans)
Figure 23 Levelling out of the number of users from 1986 (Source: RKD News March 1988, p2)
Unfortunately, it was realised that this
computer’s computing power was
insufficient. As a result, it was used
seldom and thus the IBM 4361 was
decommissioned on 1 November 1990.
The increasing use of PCs and
departmental computers resulted in
the average of the number of active
users levelling out and declining after
1986 (see Figure 23). Despite this
decline, the computer’s response
time decreased during 1989, causing
much discontent amongst users45.
Electronic mail on the mainframe
computer was announced in the RKD
Nuus of November 198948. This was
made available to all other mainframe
users at the University, South-African
universities and organisations that
were connected to Uninet, the national
academic network. Connections to
universities abroad were, in principle,
possible, but had not yet been made
available for general use. This was
the beginning of the communication
revolution created by electronic mail,
especially when it could later be used
from PCs connected to the LAN. In
comfortably house a big mainframe
computer. Provision was made for
uninterrupted power supply by
installing big batteries and a converter,
which addressed the negative effects
of unexpected power outages on
mainframe computer equipment.
Provision was also made for good
security and for firefighting equipment.
The relocation of the mainframe
computer and the accompanying
equipment could, as a result of good
planning, be completed in four days’
time during January 199049.
Figure 24 The spacious new machine room before the installation of the computer equipment. (Source: Photo collection of Mrs AD Mans)
43 Steering Committee of Computer Services, Minutes, Art 5, 7 November 1988 44 RKD Nuus, p11, March 1989 45 RKD Nuus, p6, July 1989
46 Steering Committee of Computer Services, Minutes, Art 2.2.4, 27 November 1987 47 RKD Nuus, p15, March 1989 48 RKD Nuus, p3, November 1989 49 RKD News, p1, March 1990
54 55
Figure 25 The new machine room in operation. Some of the disk drives can be seen on the foreground. The person in the background is Mrs Rika Nieuwoudt. (Source: NWU archive)
Figure 26 The batteries for emergency power in the new machine room. (Source: Photo collection of Mrs AD Mans)
Figure 27 Two examples of the IBM RS/6000 computers. (Source: Author’s photo collection)
The total disk space of the mainframe
computer was increased to 20
gigabytes in 199050. The Department
of Mechanical Engineering obtained
external funding for a powerful Convex
computer with a vector processor,
which was suitable for computing-
intensive projects. In late 1989, this
computer was installed at the Vaal
Triangle Campus’ Goodyear Building51.
In February 1990, IBM announced the
powerful RS/6000 RISC computers. Dr
Welman saw enormous potential in
these computers, and the University
was the first South African university
that ordered and received them.
Seven of these computers that were
optimised for numerical intensive
computing and required by, inter
alia, the Departments of Physics and
Statistics, were purchased52. Two of
these computers were swopped for
the IBM 9375 computer that did not
have enough computing power for
their processing needs. The IBM 9375
was then used for all mainframe-based
word processing.
The RS/6000 met the expectations
of the Department of Physics, who
used it very fruitfully53. An RS/6000
was also made available to the Faculty
of Engineering for their computing-
intensive projects, one was used for
experimental work with administrative
systems, one served as a network
server and the others were allocated to
general computer work. This resulted
in the mainframe computers from 1991
being mainly used for administrative
systems54. The large capital purchases
of 1990 unfortunately resulted in
overspending of R1.5 million, which
negatively influenced the IT budgets
for 1991 and 1992, when the shortage
had to be addressed55.
The arrival of powerful, affordable
processors like the IBM RS/6000 and
other Unix-based systems, together
with the growing number of PCs and
the development of reliable LANs,
was the beginning of the move away
from the mainframe environment. As
a result, the pressure on the IBM 4381
could be decreased and, eventually,
the University’s mainframe computer
could be phased out.
The 1980s was a time during which the
mainframe environment developed
from an environment with limited
and clumsy access to the general
interactive computer platform for the
entire University. Developments in the
late 1980s resulted in the diversification
of the computer environment with
mainframe computers remaining the
most important computer platform, but
with a variety of smaller decentralised
platforms distributing the computing
load.
50 Department of Operating Systems, Annual Report, Annexe 4, ITM, 1990. 51 Computer Services, Annual Report 1989, Art 4, in Management Committee of
Computer Services agenda, 23 November 1989 52 ITB Nuus, p12, November 1990 53 Transcription of the Author’s interview with Prof MS Potgieter, 8 September 2010 54 ITB Nuus, p25, 26, November 1991 55 ITM, Annual Report, Art 4.1.1, 1991
56 57
COMPUTER GRAPHICS
The graphical presentation of variable
values is an indispensable technique
in mathematical sciences, just as it is
a valuable resource in the geo and
commercial sciences. Thus it was
important for computer users from
these disciplines to have access to
effective equipment for the graphical
presentation of their results.
As indicated in a previous chapter, the
first graphics plotter was obtained in
197156 and the first digitiser in 197957. A
graphics plotter with a 36-inch drum was
obtained in 197958. A graphics plotter
with a 36-inch drum was obtained
in 1979 . It is not clear whether this
equipment was incompatible with the
IBM 4300 series of computers, but in the
Computer Services report dated March
1981, Prof Hattingh complains that the
graphics equipment was inadequate
and that no graphics equipment was
connected to any of their computers.
In addition, Prof Hattingh notes that
the Department also does not own a
graphics terminal, flatbed plotter or
digitiser59.
Prof Hattingh, probably as a result of
his background in statistics and a clear
understanding of the importance of
graphic presentations in disciplines like
Statistics, Physics and Mathematics,
emphasised the importance of
interactive graphics capabilities in
several presentations. The increased
pressure on the Computer Services
budget for upgrading the mainframe
computer did, however, delay the
process.
A crucial step in the direction of
interactive graphics was obtaining the
GDDM software in 198460. This enabled
graphics presentation on any computer
terminal with graphics capabilities,
something that was used far and wide.
A further step was implementing a
graphics workstation early in 1985.
It included a computer terminal, a
graphics screen and accompanying
copy unit (both from Tektronix), as well
as a plotter and digitiser (both from
Calcomp). This allowed graphics to
be presented, printed or plotted. The
graphics workstation was augmented
with an IBM 7372 6 pen plotter62.
The SAS programme suite was
obtained in 198563. It included a module
SAS Graph with powerful graphic
capabilities like three-dimensional and
other sophisticated graphics.
In the late 1980s, during one of his
visits abroad, Dr Philip Welman came
across the IBM 7350 image processor
with advanced graphics capabilities.
He contacted ISM, the local IBM
representatives about this equipment.
In the interim, ISM imported a 7350
with the hope to develop a market
for it in South Africa. This did not
prove successful, and Dr Welman
managed to convince ISM to place
the IBM 7350 at the University64. The
matter was approved by the Steering
Committee of Computer Services on
14 February 198965. Early in 1989, the
processor was installed66 and on 5 June
1989 Mr Dave Pegg from ISM officially
handed it over to Prof CJ Reinecke,
the Rector67. This garnered a lot of
interest in image processing at the
University’s Potchefstroom Campus
and a successful symposium on image
processing applications was held on 13
November 198968.
At the end of the 1980s, the University
was well-placed regarding mainframe
computer graphics capability. In the
interim, the microcomputers’ graphics
capabilities improved significantly,
especially with the arrival of the higher
resolution screens, like VGA in 1987.
Spreadsheet programmes like Lotus
123 and Quattro Pro, which had good
graphics capabilities, were quite user-
friendly. Harvard Graphics combined
high-quality graphics with easy user
interfaces. All three these programmes
were supported by Computer Services.
At the end of the 1980s, the University’s
computer graphics facilities were well-
developed.
Figure 28 The graphics workstation. A part of the digitiser is visible on the left of the photo and a part of the plotter can be seen on the right. (Source: Photo collection of Mrs AD Mans)
56 Die Wapad, Vol 26, Nr 6, p9, 1971-05-07 accessed at http://dspace. nwu.ac.za/bitstream/handle/10394/14430/Wapad_d26_n06_1971. pdf?sequence=1&isAllowed=yz
57 Newsletter, Computer Services, Art 10, November 1979 58 Computer Users Meeting, Minutes, Appendix 1, 8 October 1980 59 Computer Services Report, Art 2.1.3, April 1981, Appendix to Minutes, Steering
Committee of Computer Services, 6 March 1981 60 RKD Nuus, p15, August 1984 61 RKD Nuus, p13, March 1985
62 RDK Nuus: p23, July 1987 63 RDK Nuus, p14, March 1985 64 Transcription of Author’s interview with dr P du B Welman, 15 October 2009 65 Steering Committee of Computer Services, Minutes, Art 6, 14 February 1989 66 RKD Nuus, p17, March 1989 67 RKD Nuus, p1, July 1989 68 RKD Nuus, p24, November 1989
58 59
Figure 29 The installation of the IBM 7350 image processor in the machine room (Source: Photo collection of Mrs AD Mans)
MICRO OR PERSONAL
COMPUTERS
A lot of confusion exists around the
terms microcomputer and personal
computer referring to small single-
user computers. The term personal
computer or PC is used internationally.
As such, the term PC, in this document,
refers to the commercial multi-purpose
personal computer that is in common
use. Microcomputer refers to single-
purpose computers used for controlling
complex equipment.
The integrated microprocessor of the
mid-1970s was the predecessor of the
computer and telecommunication
revolution that entirely changed the
work and communication patterns,
e.g. the PC and cell phone that is
currently part of almost any developed
household and business. The first
commercial microprocessor was the
Intel 4004, which was introduced in
1971. This was followed by the 8008
(1972), the 8080 (1974), the 8085 (1976)
and the 8088 (1979), which formed the
basis of the IBM PC. Later, the 80186,
the 80286 used in the IBM PC/AT, the
80386, the 80486 and the Pentium
followed. The IBM PS/2’s arrival in 1987
resulted in an improvement in the PC
environment69, with faster processing,
larger memory, physically smaller disks
and better integration with other IBM
environment.
The first commercially successful PC for
the consumer market was the Apple 2
which was released in 1977. It proved
very popular and was produced until
199370. Lower-priced Taiwanese clones
of the Apple 2 became available in
South Africa in 1981 and immediately
attracted buyers, because it included an
elementary word processor, Wordstar,
and spreadsheet program, VisiCalc. It
was the first home computer for many
academics and did a lot to improve
the computer skills of University staff
members. The arrival of dot matrix
printers, like the Epson TX-80 in
197871, contributed to the increasing
popularity of the Apple clones.
An Apple 2 PC was purchased from the
Computer Services budget in 198172.
This computer was mainly used by
Computer Science students, but was
later made available to the Department
of Education Practice to determine if it
could be of use to Education students73.
Although several Apple 2 clones were
used informally by departments, large-
scale use thereof never took off at the
University. One of the reasons for this is
that Apple instituted sanctions against
South Africa in the 1980s.
Figure 30 Mr Dave Pegg (right) from ISM hands over the IBM 7350 image processor to the Rector, Prof Carools Reinecke. (Source: Photo collection of Mrs AD Mans)
69 RKD Nuus, p24, July 1987 70 https://en.wikipedia.org/wiki/Apple_II 71 Information from http://lowendmac.com/2014/personal-computer-history-the-first-
25-years/ 72 Steering Committee of Computer Services, Minutes, Art 6.1, 14 August 1981 73 Steering Committee of Computer Services, Minutes, Art 2.4(d) , 29 January 1982
60 61
Figure 31 Mr Johan Nel, Head of Operations at Computer Services at the Apple 2 microcomputer. (Source: Photo collection of Mrs AD Mans)
In 1981, the IBM Personal Computer
(later known as the IBM PC) was
released. The fact that a big and
trusted computer company’s name was
linked to the PC, inspired confidence
in consumers and, came into common
use. This computer was succeeded by
the IBM XT, which was more powerful
and included a hard drive. PCs were,
however, still expensive and the price
of the IBM XT was almost the same
of that of a medium-sized car. The
PC first made a real impact on the
University when clones of the IBM PC
and XT became available, especially
the cheaper ones from Taiwan.
Initially, sales were slow, both as a
result of a lack in computer skills and
in funds. The availability of attractive
applications like word processors did,
however, motivate people to develop
the necessary computer skills. When
prices and computer power slowly
became more attractive, a lot of people
started viewing the PC as a problem
solver worth the time and effort. As a
result, the number of PCs on campus
increased significantly in the 1980s.
There were several attempts by the
Steering Committee of Computer
Services to standardise PCs, but it was
an impossible task since there were so
many suppliers that sales could not be
controlled. Unfortunately, no central
records were kept of PC purchases
and, as such, no information on the
number of PCs on campus at the time
is available.
The only available documented
information is on the PC laboratories
that was used for student training,
which will be discussed in the following
paragraphs.
The Steering Committee of Computer
Services realised in 1986 that PC
laboratories were urgent necessities
and decided that the installation of a
PC laboratory on both campuses is
more urgent than upgrading the
overloaded mainframe computer74. At
the beginning of 1987, a PC laboratory
(with 30 Olivetti PCs) was installed on
both the Potchefstroom and Vaal
Triangle campuses.
In the beginning of 1989, two additional
PC laboratories were installed at the
Potchefstroom campus (one in the
Natural Sciences building and one in
the Library), each with 20 TSM PCs75.
The old accelerator building was
repurposed in 1990 to make provision
for the mainframe computer and several
PC laboratories. The then existing
PC laboratory was moved to the
new location. The building consisted
of laboratories with respectively 24
mainframe terminals, the 20 TSM PCs
(connected to a Novell network), 36
Olivetti PCs and 6 IBM PCs76.
The 1980s was a period during which
the PC became a commonly used
and popular resource that mainly
functioned independently. At the
end of this era, the University’s LAN,
PUKNET, was developed. This was
the first step to a totally integrated
computer environment with national
and international links between the
PCs and mainframes of academic
institutions. The installation of network-
connected PC laboratories for learning
was the next priority, since there was a
big demand for it.
Figure 32 The installation of a PC laboratory in the Ferdinand Postma library. (Source: Photo collection of Mrs AD Mans)
Figure 33 The new terminal room was one of the new spaces that were made available after the old accelerator was repurposed. (Source: Photo collection of Mrs AD Mans)
74 Steering Committee of Computer Services, Minutes, Art 5.9, 21 October 1986 75 RKD Nuus, p12, March 1989 76 RKD Nuus, p1-2, July 1990
62 63
Figure 34 One of the PC laboratories that were installed in the old accelerator. (Sour ce: Photo collection of Mrs AD Mans)
ADMINISTRATIVE SYSTEMS
The Director of Computer Services,
Mr AJ Smith, requested in May 1980
that a new department, Information
Services, be established in Computer
Services77. Mr Faan Koekemoer was the
first Head of this department, which
included Mr RH Drevin as Systems
Analyst and Mr EH van Niekerk as
Programmer. The student system was
developed from 1979. Other systems
that were developed during this era
included the ledger and leave systems.
All the systems of this era were, for the
most part, batch systems developed in
RPG2.
Considerable progress was made in
system development with the arrival
of the IBM 4331 with its VM/
SP operating system that enabled
interactive programme development.
This progress was further enhanced
by the purchase of the Focus fourth
generation language for administrative
system development. Mr A (Attie) Juyn
was appointed in 1982 as the second
programmer and he took the lead in
Focus development. In 1985, he was
appointed as Database Administrator.
The new real-time student system was
the first big system developed in Focus.
In 1984 development commenced,
and by March 1985 the heart of the
system was in use78. The first real-time
student registration was done in 1986.
The system was quite sophisticated
and could print letters to clients with
capital and lower-case characters,
as well as diacritical characters. In
1988, the student debit system was
developed and commissioned.
This system managed all aspects of
students’ financial transactions and, as
of the student registration of 198979, all
relevant functions could be performed
in real time on computer80.
The real-time telephone system was
developed in 1985, with which
extentions and user information of the
Siemens POTS was managed81. Data
from the telephone costing system
was transferred to this system to
debit departments and individuals for
telephone costs incurred.
The SANSO system that was used
to report to the government on the
University’s functioning, was developed
between 1985 and 1989 by Mr A Juyn,
Mr I Larney and Mr K Vosloo. This
system was extremely important,
since state subsidies were largely
determined by it.
In 1988, it was decided to move the
Department of Administrative
Information Systems from Computer
Services to the Department of
Management Information82. Mr EH van
Niekerk was appointed as Director of
this department and reported to Prof
Hattingh, who was then the Executive
Director of Finance and Operation. At
the time of the relocation, the systems
Figure 35 Example of a leave report from 1983. (Source: Author’s archive)
77 Steering Committee of Computer Services, Minutes, Art 4, 5 May 1980 78 Steering Committee of Computer Services, Report RKD 1st semester 1985,
Appendix to Minutes, 1 April 1985
79 Computer Services, Annual Report, Art 4 , 1989 in Steering Committee of Computer Services, Agenda, Art 5, 23 November 1989
80 RKD Nuus, p7, March 1989 81 Computer Services, Annual Report, July 1984 to July 1985 82 Steering Committee of Computer Services, Minutes Art 3.1, 29 March 1988
64 65
development staff consisted of Mr EH
van Niekerk, Mr RH Drevin, Mr A Juyn,
Mr R Coertzen, Mr J Greylingh, Mr A
Allen, Mr K Pretorius, Mrs A Bell, Mrs
E Conradie, Mrs M Janse van Vuuren
and Mrs C du Plessis, together with
seven data processors and card punch
operators.
A central students assessment system
was developed and commissioned in
198983, together with a meal payment
system. This was the first system
that made use of the client/server
configuration84.
Initially, the Focus developers
promised the inclusion of a fully-
fledged database. By the late 1980s,
however, it became clear that the
promise would not be honoured soon
and a task team was appointed to
make recommendations on a suitable
database. This did, however, only
become a reality in the 1990s.
During the 1990s, the clumsy batch
processing systems were replaced
with mainly real-time systems for the
management of the most important
business aspects of the University. This
resulted in the streamlining of many
University functions.
UNINET, THE SOUTH
AFRICAN ACADEMIC
WIDE AREA NETWORK
Dr Welman was a pioneer in connecting
the University to local, national and
international data networks. By
1982/1983, he was appointed Chair of
the Subcommittee for inter-university
networks, which was a subcommittee
of the Committee of IT Directors
appointed by the Committee of
University Heads. Other members of
the subcommittee were Mr Wille Fryer
and Mr Mike Lawrie, respectively the
Directors of Computer Services at the
Universities of Pretoria and Rhodes85.
As mentioned previously, a need
developed (independent of the work
of the abovementioned committee)
for a link between the mainframe
computers of the PU for CHE and that
of the University of Pretoria (UP). This
was driven by the Ferdinand Postma
Library’s need to make use of the UP’s
Dobis/Libis library system. This link
was established in December 1984 by
using a post office data line86. Since
both universities had IBM mainframe
computers, IBM’s protocols and
software could be used to establish the
connection between the computers.
This connection was one of the first
inter-university links in the country.
The first step of the subcommittee
for Interuniversity Networks was
connecting all South African universities
with IBM mainframe computers to one
another in 1987. The first institutions
that were connected were the PU for
CHE, the CSIR, UP and the University
of the Witwatersrand88 89.
In the same year, Uninet, the South
African academic network was
established with the help of what is
now known as the National Research
Foundation (NRF), which also served
as patron of the network. Most of the
groundwork was laid by the University
of Rhodes, the University of Cape Town
and the PU for CHE. The manager of
Uninet was Mr Vic Shaw, former IT
Director at the CSIR. The operational
work was mainly contracted to the
CSIR. Dr Philip Welman from the
PU for CHE was appointed the first
chair of the Uninet Council and
played a significant role in the further
development of Uninet. Most South
African Universities quickly joined
Uninet, but it took quite some time to
establish the necessary technology for
intercampus connections.
One of Uninet’s biggest problems
was reconciling the different network
environments of the participating
universities. As such, TCP/IP, a relatively
unknown protocol at the stage, was
selected as network protocol to
connect the different networks. In
1990 an experimental TCP/IP network
connection was established between
the PU for CHE and Rhodes University90.
It functioned very well, and TCP/IP
was used as the general intercampus
protocol.
At the time, Uninet was a South
African national connection with no
international links. Mr Mike Lawrie,
the Director of Computer Services at
Rhodes University, had links with Randy
Bush, an early internet pioneer in the
USA, who helped him to establish a
telephonic data connection with the
USA in February 1988. This connection
allowed Uninet users to send emails
to Rhodes University, from where,
with the intervention of Randy Bush
in Portland, Oregon, it was sent to
Fidonet in the USA. From here it was
further distributed internationally91. In
1991, the telephonic link was replaced
with a dedicated 9600 bps connection
that would later make international
connection much easier. In time, the
bandwidth was increased according to
demand and the financing available.
83 RKD Nuus, p17, July 1989 84 Computer Services, Annual Report, Art 4 1989, in agenda, Management Committee
of Computer Services, Art 5, 23 November 1989 85 Dr P du B Welman, The history of the development and operation of Information
Technology at the Potchefstroom Campus of the PU for CHE from approximately 1967 to 2003, contributions to Cobus van der Walt, Art 3, approximately 2009
86 RKD Nuus, p7, March 1985
87 RKD Nuus, p9, October 1988 89 P du B Welman, Report, Investigation into an academic network, Report presented
to the Inter-University Computer Committee of the Committee for University Heads, May 1987
90 Management Committee of Computer Services, meeting minutes, Art 5, 6 September 1990
91 The history of internet in South Africa, How it began, by Mike Lawrie, 1979, accessed by www2.frd.ac.za/uninet/history/ on 13 July 2017
66 67
LOCAL AREA NETWORKS
As mentioned earlier, Prof Heinrich
Herbstreith from the Karlsruhe Technical
University in Germany visited the PU
for CHE during August 1985. He was
invited by Prof JM Hattingh to advise
the University on the development of
its computer environment. As a
result of the visit, Dr P du B Welman
compiled a document that contained
the recommendations of Prof
Herbstreith, titled A Strategy for the
Use of Computers. This strategy was
approved at a meeting of the Steering
Committee of Computer Services
on 21 October 198692. This strategy
recommended that the University
develop an integrated computer
environment that made provision for a
LAN which could connect all computers
and the mainframe computer. It further
recommended that provision be
made for national and international
connections.
This strategy was purposefully
followed by Dr Welman and resulted
in investigations on how LANs could be
established and managed effectively
at the University. Mr Henry van der
Walt and Mr Ludwig Myburgh from
Computer Services did ground-
breaking work in this regard.
First, experiments were conducted
with Arcnet and, by mid-1989, the
following buildings were connected:
Administration, Law, Economic
Sciences, the Library, Frans du Toit, JS
van der Merwe and Education93. The
server resided at Computer Services.
While the network functioned well, it
was quite slow.
In 1989, experiments were also
conducted with Token Ring. This
was a network protocol that IBM
developed, and which fit in well with
the environment in which there was
an emphasis on IBM equipment. An
experimental Token Ring network
was successfully established between
Electronic Services and Computer
Services. The Library was also
equipped with a relatively large Token
Ring network94.
The Token Ring network functioned
very well and was fast. It was, however,
quite expensive and the installation
costs were high. In the interim, Ethernet
developed very well and proved to be
much cheaper. The decision was made
to migrate to Ethernet subnetworks in
the buildings.
The University’s LAN was named
PUKNET95. At the core of the LAN
was a 10 Mbps optic fibre ring with
Proteon routers in each of the following
bui ldings: Computer Services,
Physiology, Economic Sciences, the
Library and Administration. The router
at Computer Services provided access
to the mainframe computers. On the
Vaal Triangle Campus, an Ethernet-
based network was installed with
telephone data lines to connect to
PUKNET. Novell Netware was selected
as the network operating system and
PUKNET was taken into use in the
second half of 1990.
These were the first steps in the
establishment of PUKNET. During the
1990s it would expand to the most
general form of connection on the
University’s campuses, and having
national and international links.
COMPUTER-AIDED
LEARNING (CAL)
In this section, Computer-Aided
Learning at the University is not
discussed in-depth, but only where
it is mentioned in documents or
discussions about the role of Computer
Services therein.
CAL is mentioned for the first time
in the University’s documentation on
the computer environment by Prof
JM Hattingh in a report on the state
of the computing environment. This
report was compiled shortly after his
appointment as Director of Computer
Services in 1981. In this report, he
emphasises the importance of CAL
and, prophetically, states that CAL
would play an increasingly vital role in
the University, which would necessitate
considering and researching ways in
which it can be addressed effectively96.
At a meeting of the Steering
Committee of Computer Services on
29 January 198297, CAL was discussed.
This was a special meeting that only
addressed CAL in connection with
a recommendation by the Senate
Committee on CAL, that a computer
terminal and an Apple microcomputer
be placed at the Faculty of Education
92 Steering Committee of Computer Services, Minutes, Art 5.4, 21 October 1986 93 RKD Nuus, p13, July 1989 94 RKD Nuus, p13, July 1989
95 Committee for Telecommunication Networks, Minutes, Art 5.1.6, 10 September 1990
96 Computer Services, Report, Art 1.2 (i), April 1981, Appendix to Minutes, Steering Committee of Computer Services, 6 March 1981
97 Steering Committee of Computer Services, Minutes 29 January 1982
68 69
for research purposes. The matter
was carefully approached and it
was decided to place an Apple
microcomputer at the Faculty for a
semester to determine its success. No
further references to this matter could
be found.
Dr P du B Welman notes, in retrospect
on his time as Director of Computer
Services98, that the Computer Data
Corporation (CDC) Plato system for
CAL was investigated from 1982-83.
Plato terminals, which were connected
to the CDC computer in Johannesburg,
were installed at the University to
establish a test environment. There
was, however, limited interest in the
system.
In the following years, Computer
Services reported only that the lack of
computer facilities for CAL left a big
gap. In 1987, Prof C de Beer from the
Faculty of Law reported a donation
of R180 000 from Trust Bank for the
purchase of the WICAT system for
CAL99. Prof F Venter elaborates on this
in the RKD Nuus of October 1988 .
This system was quite comprehensive
and included a central computer with
storage facilities, six terminals and a
printer. The development of learning
material did, however, prove to be very
time-consuming.
From the beginning of 1987, a PC
laboratory was installed that was
available for CAL. In 1990 Dr Welman
reported that the following PC
laboratories were available at the
University101:
• Computer Science 1 30
• Computer Science 2 14
• Computer Services 1 21
• Library 21
• Economic Sciences 11
• Vaal Triangle 1 31
• Vaal Triangle 2 11
There were not suff icient PC
laboratories available, but budget
pressure was a limiting factor. During
the 1980s, facilities for CAL at the
University were slowly made available.
The reason for this slow progress was
probably, in part, a result of academic
staff’s limited access to equipment
because of the financial pressures of
the 1980s. While the installation of PC
laboratories in the late 1980s improved
the situation, it was still not sufficient.
RKD NUUS
Information provision to computer
users is an important way to
promote efficient computer
use. As mentioned, in the late
1970s regular computer user
meetings were held to address
this need. These meetings were
supplemented with several short
newsletters that were published
from 1978 to 1980. Thereafter, it
appears that communication with
computer users was interrupted for
some time.
Dr P du B Welman was appointed
as Deputy Director of Computer
Services in October 1981. He was
requested to have a newsletter
published on a regular basis.
As a result, the first edition of
the RKD Nuus was published in
August 1984. This newsletter
was a neatly printed and bound
publication, usually in A5 format
with about twenty or thirty
pages. The newsletter addressed
developments at Computer
Services and communicated
information regarding computer
facilities. For the rest of the 1980s,
the newsletter was published
two or three times per year. The
author attests to the usefulness
and thoroughness of the RKD
Nuus as a valuable resource in the
compilation of this chapter on the
1980s.
CONCLUSION
The 1980s was a period of great
development for Computer Services. It
started with a mainframe computer with
limited capabilities and that could only
handle batch processing. The decade
ends with mainframe computers that
could support campus-wide interactive
computing that was enhanced by
powerful departmental computers
and many PCs, an increasing number
of which were already campus-wide
interlinked and linked to servers and
the mainframe computers by means of
PUKNET. All these computer facilities
were also connected to other national
academic institutions. International
interactive connections would soon
become available.
Prof Hattingh and Dr Welman
established a solid foundation in the
1980s which could be further expanded
in the 1990s.
98 Dr P du B Welman, The history of the development and operation of Information Technology at the Potchefstroom Campus of the PU for CHE from approximately 1967 to 2003, contributions to Cobus van der Walt, Art 3, approximately 2010
99 Steering Committee of Computer Services, Minutes, Art 10.5, 27 November 1987
100RKD Nuus, p25, October 1988 101Executive of Steering Committee of Computer Services, Appendix to Minutes, 6
August 1990
CHAPTER 5 The Coordinating Role of the Committee for Telecommunication Networks
INTRODUCTION
The first telecommunication network
at the University was the telephone
network. Initially, a manual private
branch exchange was used, but in
the 1960s this was replaced with an
electromagnetic private automatic
branch exchange (PABX) for internal and
external telephone communication.
This became outdated in the 1980s and
parts became difficult to obtain. Mr TF
(Theuns) Schutte from the Telephone
Division launched an investigation
into replacement possibilities in 1983.
He approached Dr AJ (Cobus) van
der Walt from Electronic Services for
assistance. This was the beginning of
Electronic Services’ involvement in the
telephone network and, in general, in
the University’s telecommunication
networks.
Electronic PABXs was already available
and ISDN (Integrated Services Digital
Network) gained popularity. ISDN
was a technique by means of which
telephone exchanges could handle
voice and data simultaneously. This
implied that Computer Services should
also be involved in deciding on a new
PABX. Coordination between the
71
72 73
different departments involved proved
cumbersome. These departments
included Telephone Administration,
which managed switchboard and
telephone accounts (reporting to the
Registrar); Electronic Services, which
managed technical aspects; Computer
Services, which managed data
networks; and Technical Services that
managed the earlier load shedding
system and the civil work related to the
cable connections on campus.
ESTABLISHMENT OF
THE COMMITTEE FOR
TELECOMMUNICATION
NETWORKS
As a result of the involvement of
various departments in the University’s
te lecommunicat ion networks,
The CTN filled a big gap and
membership eventually expanded to
include nine members, namely the
Executive Director of Finance, the Head
of Electronic Services, the Executive
Director of Operations, the Director
of Finance, the Director of Computer
Services, the Director of Information
Services, the Deputy Registrar and
the Senior Electronic Technician
responsible for telephone networks.
The Committee disbanded at the end
of 1990, when the establishment of the
Department of Information Technology
and Management (ITM) took over most
of its coordinating functions. The last
meeting of this committee was held on
4 March 1991.
It is noted, with gratitude, that the
Deputy Registrar, Mr JJM (Jaap)
Verhoef, played an instrumental role
over an extended period of time
to ensure the smooth functioning
and record keeping of all the CTN’s
meetings.
LARGE PROJECTS
COORDINATED BY THE CTN
IMPLEMENTATION
OF THE ELECTRONIC
TELEPHONE NETWORK1
Purchase of the Siemens
PABX
coordination of the various aspects
became a necessity. Prof JM (Giel)
Hattingh, Director of Computer
Services, and Dr AJ (Cobus) van der
Walt, Head of Electronic Services,
approached Prof NJ (Nic) Swart, the
Vice Rector of Finance and Operations,
and requested the establishment of a
committee for telecommunications
networks. The request was granted and
the Committee for Telecommunications
Networks (CTN) was established with
Prof Hattingh at the helm, Dr van der
Walt as secretary, and Mr A (Andries)
Engelbrecht, the Registrar of Finance
and Operations, as an additional
member. The committee’s first meeting
was held on 7 June 1984.
Figure 1 Mr JJM (Jaap) Verhoef, Deputy-Registrar and manager of the telephone division. (Source: Photo collection of the NWU archive)
In the early 1980s, electronic exchanges
had already been available for quite
some time, and was in successful
operation at many universities. It was,
therefore, logical for the University
to plan for an electronic private
automatic telephone exchange (PABX).
During this time, one of the emerging
technologies was ISDN (Integrated
services digital network) that could
manage both voice and data. The
feasibility of these technologies was
investigated.
Dr AJ van der Walt and Mr PPH Schön
from Electronic Services investigated
potential service providers and
identified the following: Siemens,
AEI Henley, Philips, Plessey and
STC. Supplier information, as well as
background information from the SA
Post Office’s telephone department was
collected. There were many discussions
to determine the University’s needs.
From all the identified needs, a product
specification was compiled and a tender
document was prepared in Afrikaans
(quite unusual for the time).
Tenders were requested for PABXs for
both the Potchefstroom and Vaal
Triangle campuses. The presentations
were analysed by Dr van der Walt,
who advised that the Siemens tender
be accepted. (This happened before
the CTN was established.) This
recommendation was supported by
the Management Committee2 and the
1 Transcription of conversation between Jaap Verhoef, Pieter Schön and the Author, 10 Feburary 2017
2 Executive Committee, Minutes, Art 7.2.2.6, 12 June 1984
74 75
Assets Committee3 and approved
by the University Council on 22 June
19844. For the Potchefstroom campus,
Siemens offered two Siemens 601
exchanges to provide the requested 1
500 extensions. The two sub exchanges
were interconnected and, to a great
extent, functioned as a unit. This was
especially the case from 23 May 1988
when pulse code modulation (PCM)
interconnections were used between
the exchanges. One Siemens 601 PABX
was also provided to the Vaal Triangle
Campus.
An unusual characteristic of the tender
document was that it made provision
for a Versabraille unit to be connected
to the call registration system, to
make provision for a blind person, Mr
TF (Theuns) Schutte, to operate the
system.
Figure 2 Mr TF (Theuns) Schutte (Source: Photo collection, NWU archive)
Renewal of the
telephone cable network
There was a big demand for more
extensions on the Potchefstroom
campus. The new PABX made
prov is ion for many more
extensions than the old PABX
could handle, and this required
the implementation of a new
telephone cable network. The
old cables were mostly installed
without ducting, a practice
that was no longer acceptable.
Plans were, therefore, made for
the installation of an entirely
new telephone cable network
between buildings, which had
to be ducted before installation.
The Post Office was responsible
for the cables and casings, but
the University was responsible for
digging the necessary trenches
for the ducting. It was attempted
to dig the trenches in a very short
time, which disrupted gardens
and pedestrian traffic for quite
some time. The ducting that was
installed played a key role in the
development of comprehensive
data and management networks
at the University.
Exchange connections
One of the limitations that had to be
managed was the limited number of
exchange connections that the Post
Office could provide between the PABX
and the public exchange. This was a
problem at both campuses, but was
more relevant to the Potchefstroom
campus with its much higher telephone
traffic. A partial solution was to
integrate in the new PABX most of
the existing exchange connections to
private departmental numbers. The
problem was only solved when, in
April 1985, the Post Office upgraded
the public exchange and provided
sufficient exchange connections. The
final solution was presented in 1986,
when an electronic public exchange
for Potchefstroom was installed and
z-screen cables were used to provide
PCM connections to the campus.
At the Vaal Triangle Campus, the
number of exchange connections was
also a limiting factor until a new public
exchange was installed at Bedworth
Park in 1988, and the campus could be
connected with a z-screen cable.
Planning the new telephone
network
With more extensions than ever before,
the redesign of the telephone network
and awarding of additional extensions
was necessary. Almost all the existing
telephones would have to be replaced
with DTMF telephones for faster
dialling.
Dr van der Walt developed a new
telephone database in Focus, which
could manage the applications of
departments for extensions, the
awarding thereof and the submission of
the applications for the new extensions
to the Post Office. This system could
also print the internal telephone guide.
The Post Office pushed for the timeous
submission of installation applications,
since it would require a lot of work from
their side. The database was, therefore,
completed in a short time.
Special mention must be made of
the excellent support provided by
Mr AL (Bert) Venter, Head Technician
of External Works of the Post Office
in Potchefstroom. He was especially
accommodating to the University
during the renewal of the telephone
and cable network of the Potchefstroom
campus.
3 Asset Committee (Finance sub-committee), Minutes, Art 4.10, 22 May 1984 4 University Council, Minutes, Art 9.2, 22 June 1984
76 77
The commissioning of the
new telephone exchanges
Everything went more or less according
to plan with the commissioning of the
new PABXs. The exchange at the Vaal
Triangle Campus was commissioned
on 26 October 1984, followed by that
at the Potchefstroom campus on 24
November 1984. Both exchanges
functioned very well.
Improvements to the PABX happened
systematically. Direct dialling to the
extensions was implemented at the
beginning of 1987 at the Potchefstroom
campus, and on 13 July 1990 at the Vaal
Triangle campus.
The Siemens 601 PABXs remained
in operation for 20 years, until 2004.
This attests to Siemens’ technical
expertise to remain competitive with
an exchange that used an Intel 8086
processor (the same as used by the
Olivetti PCs) and 2-megabyte memory.
It is interesting to note that most of the
Siemens 601 development was done
by Siemens engineers in South Africa.
The University entered into an
agreement with Siemens South Africa
to allow the University to themselves
maintain the PABXs, which saved a lot
of money in terms of maintenance.
Figure 3 Mr PPH (Peter) Schön
This required that a new position be
created, namely that of Senior
Electrical Technician in the Department
of Electronic Services. Mr PPH Schön
was appointed in this position, where
he served in related positions for more
than twenty years. He was responsible
for maintaining the Siemens exchanges
and was recognised as an expert by
Siemens South Africa, who often
consulted him.
Telephone cost
management
The rental of telephone extensions and
cost of telephone calls were initially
covered by the central budget. Limiting
these costs became increasingly
diff icult until the budget was
distributed amongst the departments
in 1986 and they had to manage their
own telephone budgets. This lead to a
significant cost saving and a reduction
in the number of extensions used.
Connection of hostel
telephones to the PABX on
the Potchefstroom campus
For a long time, telephone traffic to
on-campus student residences was
handled by small manual exchanges
in the hostels. Incoming telephone
traffic was very high, especially during
the evenings, making it rather difficult
to make telephone calls to students
in a residence. The problem was
solved from 1988 on by connecting
the telephones for incoming traffic to
the student residences, to the PABX.
The incoming exchange connections
to the PABX were underutilised after
office hours and therefore available for
student residences.. This successfully
solved the communication bottlenecks
to the student residences and resulted
in great user satisfaction.
Connecting the hostel telephones to
the central PABX had an interesting and
beneficial additional advantage. During
a visit by the Postmaster General, Mr
S du Toit Venter, to the University, he
was requested to assist the University
regarding the high cost of rental
of external telephone extensions.
Connecting the student residence
telephones led to a significant increase
in calls and thus in income to Post
and Telecommunications, due to the
substantial higher number of exchange
connections that were available in
the evenings for calls to residences.
Mr Venter agreed that the cost of
rental for the external extensions
would be the same as that for internal
extensions5. As a result, the rental cost
of departmental extensions decreased
by approximately a third from April
1995, campus-wide. This resulted a big
annual saving.
5 ITB Inligting, p1, April 1995
78 79
IMPLEMENTATION OF THE
LOCAL AREA NETWORK
The growing number of PCs on the
campuses made interconnections
between them necessary. In addition,
management systems like the meal
payment system required network
connections for the relevant PCs. PC
networks were managed by Computer
Services, while technical side of the
meal payment system was managed
by Electronic Services. Although the
planning and implementation of
PUKNET was mostly done by Computer
Services, the CTN coordinated
and partially financed the process.
The implementation of PUKNET is
discussed in detail in Chapter 4.
DIGITAL UNIVERSITY
ACCESS CARDS
Requests were made by several parties
for the implementation of digital
university access cards. The motivators
were especially Security Services for
access control and Student Residence
Administration for implementing a
meal payment system. An important
requirement was that card readers had
to be weather proof and capable of
functioning at campus access gates. An
investigation indicated that Wiegand
access cards were the best solution6.
The advantage of the Wiegand access
cards is that they are difficult to forge
and that there were waterproof card
readers available for them. Electronic
Services designed the coding of the
access cards and provided a check
character to minimise reading errors.
A local supplier, who had the access
cards manufactured in Germany, was
identified. When the access cards were
delivered early in 1988, it appeared
that they were erroneously coded. The
urgency to release the access cards at
the beginning of the academic year
did not allow for the access cards to
be returned to the supplier. Electronic
Services could address the erroneous
coding by decoding the access cards
using a very complex algorithm. This
had the unexpected advantage of the
access cards being even more difficult
to forge.
The first large-scale access control
system was developed for the Ferdinand
Postma Library and was commissioned
during the July holidays of 1988. The
next project was the basement door
of the Administration building and the
access gates at the western entrance of
the campus. Thereafter, several access
control systems were implemented.
The digital University access cards
enabled the development of a meal
payment system for hostels, which
had to be put into commission
at the beginning of 1998. Hostel
administration compiled specifications
for the system, requested tenders
and appointed a contractor. A week
before the students arrived for the new
academic year, the contractor admitted
that he was incapable of delivering
the system. Then Mr EP Blignaut of
Electronic Services designed and
implemented an emergency system
within 10 days7. This system, with its
updates, formed the foundation of the
meal payment systems for the next ten
years.
LOAD SHEDDING SYSTEM
The costs of electrical energy at a big
organisation is determined by energy
use in kilowatt hour (kWh) and the
maximum demand in kilovolt ampere
(kVA). To decrease energy costs for
the University, it became necessary
to control both components. The
maximum demand could be controlled
by switching off large loads, like big
geysers in hostels, during periods of
high demand.
Technical Services managed a
Honeywell load shedding system from
1985. This could switch off a number
of geysers in student residences
during peak times. The system was,
however, limited in its capabilities. The
increasing energy costs required a more
comprehensive load control system.
Electronic Services planned the system
and purchased a commercial load
control management software system.
This enabled the University to monitor
and intelligently control all the large
geysers to minimise the maximum kVA
demand. In February 1990, the system
was commissioned and resulted in
significant electricity cost savings. This
system was coordinated and financed
by the CTN.
The load management system was
later on, when the new department,
ITM, was instituted, the responsibility
of the networks division of ITM.
CONCLUSION
Before the format ion of the
department of Information Technology
and Management, the Commission
for Telecommunication Networks
successfu l ly coordinated the
development and management of the
important communication networks
of the university. From 1991, this
coordination could, to a large extent,
be managed by ITM.
6 Committee for Telecommunication Networks, Minutes, Art 5.1, 12 November 1987 7 Electronic Services, Annual Report, 1989
6
CHAPTER
Library computerisation
INTRODUCTION1
As the guardian and retriever of a large
amount of academic information, it
goes without saying that the Ferdinand
Postma Library as a university library,
would have to utilise information
technology. Appropriate affordable IT
resources were not available in the early
years. However, the University Librarian
since 1976, Mr (later Prof) CJH (Casper)
Lessing, was aware of the possible
applications of information technology
in libraries, and already started using
it in the 1970s. Later, it was exploited
and used more intensively, to such
an extent that the Ferdinand Postma
Library became a leader in certain
fields in South Africa and abroad. From
1980, Prof Lessing was supported in
this venture by Mr (later Dr) T (Tom)
Larney, who later became the Assistant
Director (Research and Development)
at the library. He was computer literate
and could develop computer software.
1 This chapter is mainly based on information from transcriptions of interviews conducted by the Author with 1) Prof CJH Lessing and Dr T Larney on 3 March 2010, 2) Dr T Larney and Mrs C Breytenbach on 16 February 2010 and 3) later private correspondence with Dr T Larney.
81
82 83
Figure 1 Prof Casper Lessing Figure 2 Dr Tom Larney
FIRST APPLICATIONS
The Ferdinand Postma Library
purchased several programs from the
British Library in the mid-1970s, with
which information on new bibliographic
records, which were distributed by
British Library on magnetic tape, could
be selected and printed. This was very
useful in bringing new publications
to the attention of the academic
personnel.
By the late 1970s, the digital
bibliographic records of library
resources were captured in the
internationally-recognised MARC
format. Electronic Services designed
customised digital equipment
for this procedure. Capturing
digital bibliographic records using
computerised equipment was in
general use until such time that the
library would have an integrated
computerised library system. Some
of the information that was selectively
extracted from the abovementioned
British Library tapes, was also used2
for this purpose.
INTEGRATED LIBRARY
SYSTEMS
Integrated library systems are used to
computerise the important operational
functions of a library, such as purchase,
cataloguing, lending, information
supply, etc. Early on, the library was
already aware of the possibility of
using the German/Belgian Dobis/
Libis system for integrated library
management. This was a powerful
system that, in principle, could run
on the University’s IBM mainframe
computer, but demanded much
processing power and storage capacity.
In the financially-difficult 1980s, there
were insufficient resources available
for the University’s computer to run
the Dobis/Libis system. A module
was, however, purchased for capturing
bibliographic records. This allowed
the capturing of bibliographic data to
continue. From 1985 onwards, the link
between the PU for CHE’s mainframe
computer and that of the University of
Pretoria, which already used the Dobis/
Libis system, was used. This enabled
the Ferdinand Postma Library to extract
information from the UP’s Dobis/Libis
system and explore the system. By
1988, a number of modules from the
Dobis/Libis system was implemented
on the local mainframe computer3. The
system was, however, never completely
implemented. By the end of the 1980s,
2 Computer Services, Annual Report, Art 8, 1986, Appendix to minutes, Executive Committee Computer Services, 21 October 1986
3 Steering Committee, Computer Services, Minutes, Art 9, 1987-11-27
84 85
it became clear that the Dobis/Libis
system reached the end of its useful
lifetime and was terminated in 1991.
This necessitated the selection of a
new integrated library system.
After a thorough study of several
relatively unsatisfactory systems, it
was decided to purchase the Data Trek
system from Data Trek Inc in the USA.
This was a DOS-based system that
could run on a computer network. It
was successfully commissioned by 1993
and was of great advantage to the
Ferdinand Postma Library, since it was
a completely integrated system that
could manage most library functions.
Amongst other advantages, this meant
that the massive card catalogue could
be removed from the library to make
space for modern computerised
resources.
Although Data Trek provided
excellent service, a more attractive
option became available when the
GAELIC consortium of South African
universities and Technikons north of the
Vaal River was formed. This consortium
successfully negotiated financing for a
comprehensive and integrated library
system for all tertiary institutions within
the GAELIC consortium, with the Mellon
Foundation in the USA. In 1997, GAELIC
decided to purchase the Innopac
system from Innovative Interfaces Inc
in the USA. All cooperating tertiary
institutions’ bibliographic records had
to be adapted to the new system. The
Ferdinand Postma Library’s continuous
and consistent use of the international
Marc format for bibliographic records
proved very beneficial, since they
could import their bibliographic
records into the new system without
the assistance of external contractors.
The Innopac system was implemented
and successfully used. Innopac
was upgraded in 2001 to include a
graphic interface, and was known as
the Millennium system. It rendered a
valuable service to the Library.
ELECTRONIC MAGAZINE
INDICES AND
INFORMATION DATABASES
Quite early on, Prof Lessing and his
colleagues became aware of the
importance of bibliographic indices of
academic journals to academic
users. Apart from printed bibliographic
records that they had already made
available in the 1970s, they subscribed
to the DIALOG digital information
service in the USA in 1981. Initially, this
service was connected to by means
of a telephonic connection using an
acoustic modem. Later, the cheaper
and faster packet-switching network
of the SA Post Office, Saponet, was
used. DIALOG was an extremely useful
service, something which the author
could attest to.
DIALOG was the predecessor of
the computerised magazine indices
which the university acquired and
which made it a pioneer amongst
South African universities. This started
with the purchase of their first CD-
ROM, the Grolier Encyclopaedia, in
1986. This was very successfully used
and resulted in more CD-ROMs with
magazine indexes and other databases
being purchased. Pioneering work
was done to extract this information
in an as yet technologically unstable
environment. Soon CD ROM readers
were interconnected by means of the
CD-Net network, which simplified
access to information. Further
development enabled the CD ROM
readers to be connected to PUKNET
from approximately 1990. This
technology was further developed, in
particular with the HP SureStore multi-
CD ROM drivers, which could connect
a large number of CD ROM drives to
the network. There were approximately
200 CD ROM drivers in this system,
making it one of the biggest CD ROM
networks in the world at that time.
Although this was a very important
and useful facility, it was only an interim
phase before all magazine indices and
full text databases would be made
available on the World Wide Web by
the mid-1990s. These information
sources were later made available at
a reasonable price by consortiums like
GAELIC. Commercial companies, like
EBSCO and SWETS, also made many
information databases available at an
affordable price through the Internet.
The Ferdinand Postma Library
developed a website for the library in
1996, which made information sources
accessible via the web. This system
was expanded from 2000 onwards to
provide access to many information
databases and the Innopac library
system, on-campus and also for off-
campus students.
It is clear that the Ferdinand Postma
Library was successful in providing
excellent service in computerised
access to academic journals and
information databases for academic
users.
CONCLUSION
At the end of 1998, Prof Lessing
retired after a very fruitful service of
four decades, during which a lot was
accomplished to make academic
information available in an effective,
computerised manner to academics
and students. His innovative work
methods were continued by his
successors, Mr PE (Paul) Buys and Mr
(later Dr) T (Tom) Larney. They continued
and enhanced the excellent service to
both academics and students through
computerised methods.
CHAPTER 7 Information Technology and Management, 1991 to 2003
INTRODUCTION
In the 1980s, information technology
was characterised by the domination of
the mainframe computer. This changed
in the 1990s when servers and networks
dethroned the mainframe computer
as the king of the computer world.
In fact, as part of this movement, the
University’s mainframe computers were
decommissioned in 1996. From then
on, specialised servers and networked
PCs with ever-improving national and
international networks, would take the
lead.
The management of information
technology at the University also
changed at the start of the 1990s.This
chapter covers the time period from
the formation of the new Department
of Information Technology and
Management (ITM) in October 1990
until the PU for CHE’s end as an
independent entity in 2003.
Important factors determining the
strategy of ITM included the need for
effective University business systems,
87
88 89
the change-over to the new millennium,
the broad expansion of the data
communication network and, later on
the need for student support systems,
learning management systems and
support for distance education.
Figure 1 T1 The Vice Rector, Prof Albert Viljoen (left) to whom the Director of ITM, Dr Cobus van der Walt (right) reported to from 1992 to 2000. This photo was taken in June 1998, during the welcoming of Dr Michael Zastrocky, Vice President of the Gartner Group, which were leading IT advisors to ITM.
THE ESTABLISHMENT
OF THE DEPARTMENT
OF INFORMATION
TECHNOLOGY AND
MANAGEMENT
The Universi ty ’s Management
Committee established a commission
on 23 January 1990, which comprised
Prof CJ van der Watt, Chief Director of
Staff; Prof P van Eldik, Executive Director
of Technology and Development; and
Prof JH Venter, professor in Statistics
and a demanding computer user, to
advise about the optimal structure
within which the computer-related
departments at the PU for CHE could
be organised to form an effective
information technology support
structure. Because the University had
become strongly dependent on this
technology1, this new entity had to
provide effective advice, planning and
operational functions in information
technology2.
The commission consulted far and
wide and took note of the Committee
for Telecommunications Networks’
recurring requests for the
establishment of a division of
Telecommunication Services in the
same structure as that of Computer
Services. The first report presented
to the Management Committee on
8 May 19903 was accepted, and the
final report was presented to the
University Council on 5 September
1990. The University Council accepted
the commission’s recommendations
on 28 September 1990. It was decided
that the departments of Computer
Services and of Operational and
Information Services, as well as the
Telephone Division that formed
part of Meeting Administration, be
combined to form a new department,
namely the Department of Information
Technology and Management (ITM).
This Department would report to the
Executive Director of Technology and
Development, Prof P van Eldik. Dr
AJ (Cobus) van der Walt, previously
the Head of Electronic Services, was
appointed as Director of the new
Department, effective 1 October 19904.
His first task was to structure the new
department.
ITM initially consisted of five divisions:
Operations, Operating Systems,
User Training, Information Systems
and Networks. The Department of
Electronic Services was soon added
as an additional department, and
Administrative Support operated
as a staff function. ITM Vaal Triangle
Campus was initially incorporated into
User Training, but was later added as
a seventh ITM division. The structuring
of ITM had to be managed within the
limited human resources budget of
1991. There was, however, latitude to
correct a few historical deficits, and
small salary increases were given to
most employees.
The first step was to finalise the job
descriptions of the managers of
the divisions, and to do job grading
according to the Peromnes system.
Clearly vacant and new positions were
advertised and successful applicants
were appointed. Divisional managers
were appointed, effective 1 November
1990. Thereafter, positions within the
divisions were structured and the
necessary appointments were made,
effective 1 December 1990.
GOVERNANCE,
MANAGEMENT AND STAFF
Governance
All important decisions regarding
ITM and the computer environment
had to be approved by the Executive
Committee. Especially important
decisions also had to be approved
by the University Council. An attempt
was made to provide the Management
Committee with core information
regarding the strategic importance
of information technology during
the annual visits of the Management
Committee to ITM. Later in the
decade, an effort was made by the
Director of ITM to conduct regular
visits to members of the Management
Committee. This created a better
understanding of the possibilities and
problems of information technology.
1 CJ van der Watt, Report regarding restructuring and development of Information Technology at the PU for CHE, 5 September 1990
2 Management Committee, Minutes, Art 3.2, 1990-09-11
3 Management Committee, Minutes, Art 33, 1990-05-08 4 Management Committee, Minutes, Art 3.2, 11 September 1990
90 91
The ITM Director initially reported to
the Executive Director of Technology
and Development, Prof Peter van Eldik.
After the latter left the employ of the
University in 1991, ITM reported to the
Vice Rector, Prof AJ (Albert) Viljoen. The
position of Vice Rector of Technology
was established at the beginning of
2001; ITM then had to report to this
position.
The Management
Committee’s annual visits
to ITM
The Management Committee
conducted annual visits to ITM from
1992. These visits of a couple of hours
allowed ITM to inform the Management
Committee of strategic developments
in information technology. The topics
that were discussed over the years
give an overview of the change in the
strategic vision of the ITM Director and
managers. (Unfortunately, not every
year’s programme of the Management
Committee’s visits to ITM are available.)
During the first visit in 1992, an
overview was given of the structure
and funct ioning of the ITM
department, including a strategic
vision of expected developments
in information and communication
technologies. The significant impact
of the telecommunications revolution
was pointed out as well. The old
paradigm where a user was linked to
a centralised mainframe computer
was being replaced by the experience
that the user was in the centre of an
information universe. This user is
served by a local area, wide area and
international networks. All of this would
create a multitude of new possibilities
in distance education5.
In 1993, emphasis was placed on the
technology vision for 1997-2000 and
the planning for the decommissioning
of the mainframe computer6.
In 1995, the focus was, amongst
others, on the fact that effective
management and operation of
information technology had become
a critical success factor for universities.
Emphasis was also placed on the
fact that the world of information
technology was becoming increasingly
complex. During this year, the way was
also paved for the appointment of a
permanent Deputy Director of ITM7.
In 1998, the transition to the new
millennium was prioritised.The
increasingly strategic role that IT-
empowered organisations would play
in the future, which would require a
transformation in the way universities
were considered by the corporate
world8, was also accentuated.
In 1999, it was pointed out that
information technology had become
a part of our way of living. In the
commercial sector, it has already
become a differentiator between
successfu l and unsuccessfu l
organisations, which also increasingly
applied to universities. The importance
thereof had to be considered in the
University’s strategic, operational and
staff planning. The global availability of
IT staff was a growing problem that also
affected South Africa. The deployment
of information technology had to be
done in such a way that it did not
unnecessarily increase complexity so
that staff requirements remained within
reasonable limits. One of the most
important ways to limit complexity was
the standardisation of software and
equipment. Focus was also increasingly
being shifted to the nearing transition
to the new millennium9.
In the year 2000 (commonly referred to
as Y2K), focus was shifted to the impact
of the coming tsunami of electronic
business (e-business) and mobile
business (m-business) on organisations
and how the University should prepare
itself for the future. The Rector, Prof
Carools Reinecke, requested that Dr
Cobus van der Walt gather a team
of experts from the University to
compile a strategy and business plan
on the implementation of the virtual
aspects of flexible learning at the PU
for CHE. An agreement was reached
on a deadline of three weeks after the
meeting. This resulted in a strategy
document titled Strategy and business
plan for implementing the virtual
aspects of flexible learning at PU for
CHE 10, referred to later on as the Three
Weeks Document.
Coordination and advisory
committees
With the establishment of ITM, the
old Steering Committee of Computer
Services ceased to exist. In its place
the Coordination Committee ITM and
the Advisory Committee ITM were
formed. The Coordination Committee
served as a forum for the large users
of ITM facilities and services to state
needs and exchange information. The
Advisory Committee was a very useful
forum in which several experts and
equals from other universities and the
private sector could comment on the
strategy and facilities of ITM.
5 IT-geskiedenis\Winword\1992 BKbesoek.doc 6 IT-geskiedenis\Winword\1993 BKBESK_2.doc 7 IT-geskiedenis\Winword\1995 BKBSK ITB gesprek met Bestuurskomitee 24 Okt
1995.doc
8 IT-geskiedenis\Winword\1998 BK98CW.ppt 9 IT-geskiedenis\Winword\1999 BK besk beklemtoonde sake.doc 10 van der Walt, editor, Strategy and business plan for implementing the virtual
aspects of flexible learning at PU for CHE, 2000, IT-geskiedenis\Winword\ DriewekedokumentRektorDinsdag.doc, Maart 2000
92 93
ITM directors
(This subsection was written by Mr
Boeta Pretorius, the current (2017)
Executive Director of the North-West
University’s IT Department.)
In October 1990, Dr Cobus van der
Walt took over the reins and served
as the ITM Director for just over ten
years. During these years it became
increasingly important to focus on
the client and, as such, the Customer
Service Department was established.
For the first time, requests and
problems were managed by a system.
One of the biggest challenges of the
late 1990s was planning the transition
to the new millennium. Dr van der Walt,
however, guided the Department to a
problem-free transition.
One of the sustainable successes of
this era was the establishment
Figure 2 Dr AJ (Cobus) van der Walt, ITM’s first Director (Source: Author’s photo collection)
of the company IT3-Educational
Services, which was and still is, in
2017 (OpenCollab) responsible for
administrative system development for
the University. The Forte development
system for administrative systems was
purchased and successfully used for
many years.
One of the highlights of Dr van der
Walt’s tenure and something that most
remember him for, was the beginning
of ITM’s direct involvement in virtual
learning. The Rector requested that
he develop a document on the use
of technology in education at the
University; he was given three weeks
to complete this document, which
became known as the Three Weeks
Document and is still referenced today.
Dr van der Walt handed over the
Directorship to his Deputy Director, Mr
Attie Juyn, at the end of 2000.
Figure 3 Mr A (Attie) Juyn was ITM Director from 2001 to 2003 and Executive Director of Information Technology from 2004 to 2014.
Mr Attie Juyn took over as Director in
2001, where he served until the end
of 2014 when he accepted a position
at the University of Stellenbosch.
Although he had a major influence on
the expansion of the ITM department,
this section will only mention a few
highlights between 2000 and 2003. Mr
Juyn had a background in business
systems and played a significant role
in the planning and building of a new
student system. He firmly believed in
cooperation with other national and
international universities and played
a significant role in South Africa’s
involvement in two international
projects, namely Kuali and, to a lesser
extent, Sakai. This period was marked
by preparation for the merging of the
three campuses to form the North-
Figure 4 ITM management in 1996. From left to right in front: Mr Ernst Blignaut, Manager of Electronic Services; Dr Cobus van der Walt, Director of ITM; and Mr Richard Drevin, Operations Manager. Back: Mr Ernest van Niekerk, Manager of Information Systems; Mr Jan van Rooy, Manager of Operation Systems; Mr Attie Juyn, Information Technologist; Mr Wilhelm van Belkum, Network Manager; and Mr Boeta Pretorius, User Training Manager. (Absent: Mrs Aldine Oosthuyzen, Manager of ITB Vaal Triangle campus)
West University. The extent of IT’s
centralisation or decentralisation was
not technically demanding, but
emotionally challenging.
ITM management
ITM management consisted of all the
managers of the different ITM divisions.
The members of the ITM management
of 1996 are depicted in Figure 4.
The divisional managers had more or
less full authority over their divisions.
The authority, responsibility and
accountability of the Director and
Managers were defined by a
management matrix that was agreed
upon between them11.
11 IT-geskiedenis\Winword\Bestur13 Bestuursmatriks van ITB.doc, 2000-10-30
94 95
Subdivisions and staff
members of ITM
As previously mentioned, there were
seven divisions in ITM in 1991. The
following table depicts these divisions
and their initial staff complements:
Department
Number
of staff
members
Administrative Support 4
Operations 7
Operational Systems 3
Electronic Services 4
Information Services 8
User Training 8
Networks 10
Vaal Triangle Campus 4
Total 48
Initially, there were 49 staff members in
ITM (Director included).
One of the problems at the start of
ITM was the physical location of the
department’s divisions of Information
Systems and of Networks being in
other buildings than the rest of ITM.
Information Systems was located in the
Administration building, and Networks
was located in the Physics wing of the
Natural Sciences complex. Electronic
Services was also located in the Physics
wing but due to the special building
requirements for Electronic Services,
it was not economically justifiable
to move them. By the end of 1995,
Computer Sciences moved to other
floors in the same building, which made
more space available for the divisions
Information Systems and Networks to
move to the same wing as the rest of
ITM. This helped to create a better
sense of unity amongst the ITM staff.
In time, the staff members and needs
changed, which resulted in restructuring
the divisions and decreasing the
number of them. The department
of User Training and a section of
Operations (Helpline and PC-storage)
was merged in 1997 to form the
division of Customer Services, which
was headed by Mr Boeta Pretorius. The
establishment of this division illustrated
the growing emphasis being placed
on the customer. Prof EJ Smit, retired
University Registrar, was appointed the
first Head of the Helpline in Customer
Services. The Remedy Action Request
System was commissioned in 1997
to keep track with user requests. The
machine room responsibilities were
transferred to Networks12. With the
resignation of Mr Jan van Rooy in 1997,
the Operating Systems’ functions were
divided between Customer Services
(applications) and Networks (UNIX
environment operating systems). From
1 March 1999, Electronic Services were
separated from ITM to form part of
Specialist Academic Services.
ITM was structured as follows as of
1 March 199913:
The ITM structure was adapted again
in 2001 when the Customer Services
team’s expansion required that it be
divided into two divisions, namely End-
user Technology with Boeta Pretorius
as Manager, and Helpline, with Daleen
Botha as Manager14.
In October 2004, a set of photos were
taken of the IT staff complement, which
is similar to the staff complement of
ITM at the end of 2003, the last year
included in this history.
Figure 5 The new department, Customer Service, in 1998 after a restructuring of divisions in ITM in 1997. Boeta Pretorius, the Customer Services Manager, stands at the right of the second row and Prof Rassie Smit, Helpline Manager, is second from the left in the first row. (The ITM Director and D e p u t y Director are also present on the photo, but were not members of the division.) (Source: Photo collection of the NWU archive)
12 A Juyn, Customer Service, Proposal for restructuring, 24 January 1997, IT- geskiedenis\Winword\KlienteDiensVoors.doc
13 ITM, Annual Report, Art 2.4.2, 1998 14 ITM Annual Report, Art 2.2, 2001
96 97
Figure 8 ITM Information Systems: Front row, from left to right: Phillip du Plooy, Ida Richter, Marinda Riekert, Mari Prinsloo, Anna-Marie Bell, Ernest van Niekerk (Manager). Back row: Jan- Louw du Preez, George Gericke, Johan Badenhorst, Johan Greyling, Le Roux Kruger
Figure 6 All the staff members of ITM.
F i g u r e 7 ITM Management: From left to right: Rika Nieuwoudt (Customer Services Manager), Wilhelm van Belkum (Networks Manager ) , A ld ine Oosthuyzen (Manager of ITB Vaal Triangle campus), Ernest van Niekerk (Manager of Information Systems), Attie Juyn (Director), B o e t a Pr e t o r i u s (Manager User Support)
Figure 9 ITM User Support Front, from left to right: Chrissie Blume, Adelle Lotter, Géna Nel. Middle, from left to right: Paul Buys, Wilna Oosthuizen, Gerda Schilling. Back, from left to right: Boeta Pretorius (Manager), Petra Riedel
98 99
F i gu r e 10 ITM Customer Services Front, from left to right: Ivano van Rooyen, Philna Barnard, Hettie Jonker, Koos Brits, Geraldine Gerhardi, Alta van Zyl, Thabo Molambo. Middle, from left to right: Ria van Deventer, Rika Nieuwoudt (Manager), Dirk Venter, Barend Cilliers. Back, from left to right: Heinrico Bekker, Chr is tof f Welthagen, Gawie le Roux , Johan Swanevelder, Thys Snyman, Flip van Schalkwyk, Riaan van der Walt
F i gu r e 11 ITM Networks Front, from left to right: Wilhelm van Belkum (Manager), Stef Bezuidenhout, Engela Wilson, Hannes Kriel, Peter Schön, Lappies Labuschagne, Mathews Seroalo. Back, from left to right: Dawie van den Berg, Wim Rankin, Riaan Marx, Jopie Dulhardt, Leslie Muller, Pieter Enslin, Sias Makgale
BUDGET APPROACH
One of the biggest influences on the
functioning of ITM in its early years was
the shortage of funds as a result of the
amount of R1,44 million overspent
on the capital account of Computer
Services by the end of 1990. This,
together with the last payment of R380
000 on a loan by Senbank to Computer
Services and R300 000 worth of unpaid
invoices at the end of 1990, resulted in
very limited funds for capital purchases
during 1991 and 1992.
The current account was also under a
lot of pressure as a result of the high
annual software and maintenance
costs of the mainframe computers
(approximately R1 million per year). In
addition, the mainframe computer’s
printing expenses amounted to about
R220 000 annually. The running costs of
the mainframe computer in addition to
the mainframe printing expenses and
data line costs took up about 75% of
the ITM’s operational cost allocation. A
further problem was that the mainframe
computer costs was USD-based and,
with the ZAR continually declining in
value, annual operational costs
continually increased. (The ZAR value
versus the USD declined by about 33%
from 1991 to 1996.)
Several approaches were followed to
find a solution to the budget pressure.
A decision was taken to
phase out mainframe computers in
the long run, but this required that
alternative solutions be found for the
University’s business systems. Finding
good alternatives for these business
systems was a time-consuming and
arduous process that took up a lot
of time during the 1990s (this will be
discussed in more detail later in the
chapter).
Another problem with financing was
the earlier practice that the Computer
Services budget had also to provide
for computer equipment in the
academic departments. This resulted
in a stressed relationship between
Computer Services and the academic
departments. For this reason, ITM
requested approval for a different
budget approach that allocated a part
of the ITM budget to deans, only to be
used for computer equipment. Every
dean then had to distribute that budget
between the academic departments in
the faculty of the relevant dean. This
also applied to printing costs and disk
space, for which cost recovery was
instituted from September 1991. These
arrangements resulted in a smaller
budget for ITM, which only had to
provide central facilities.
100 101
THE UNIVERSITY’S
BUSINESS SYSTEMS
Migration from the
mainframe computer
As mentioned earlier, real-time systems
for the University was developed
with the Focus fourth-generation
development environment during the
1980s. By 1989, the student system and
the central students assessment system
was already developed as real-time
systems and development of systems
that would be of strategic importance
to the University was continued. The
mainframe computer’s slow response
time did, however, delay system
development.
As a result of the decision to phase
out the mainframe computers over
the long-term, an alternative computer
platform had to be found for business
systems. One of the possibilities
under investigation was migrating to
a package environment. Here, the ITS
package of the South African company,
Integrated Tertiary Software (Pty) Ltd,
was investigated in-depth, as per a
decision made during a meeting of
the ITM Coordination Committee on
1991-07-0315. The user departments
reported at the following meeting on
1991-08¬-2216 that the ITS systems did
not in general address their stated
needs. At the meeting held on 1991-
12-0217, it was decided to abandon this
avenue of investigation.
The other possibility that was
investigated was migrating the Focus
systems to a Unix platform, like the
IBM RS/6000, the operational costs
of which was approximately a third
that of the mainframe environment.
Benchmark tests were performed on
RS/6000 computers at IBM in München,
Germany, which showed positive
results. In a test run, the warehouse
system was successfully migrated to
the RS/6000 environment in 1992.
It was decided that the client/server
environment was a better architecture
to use. In 1993, the human resources
system was migrated to the client/
server architecture using the IBM
RS/6000 as a server. This proved
successful, but several problems
were experienced with the Focus
database, which could only be solved
after a visit abroad by Mr Attie Juyn,
ITM’s information technologist. There
was a vague discomfort over the
long-term use of Focus as a data
base environment, since it became
clear that the Focus developers paid
relatively little attention to the actual
database part of their development
environment18.
A problem with migration to the client/
server environment, was the amount
of time it would take to redevelop the
financial systems. A lot of discomfort
was experienced with the present
system as a result of the lack of
commitment accounting and the long
waiting period for ledger reports, and
a quick solution of this problem was
desired.
All of this this resulted in a penetrating
investigation of the strategy for the
University’s administrative information
systems.
Inter-university cooperation
in business system
development
The possibility of interuniversity
cooperation for system development
was investigated. This matter was
presented by Prof JT Meij from the
University of Stellenbosch (US) at a
meeting of Directors of the Information
Technology Departments of the South
African Universities in 1994. Such a
strategy was investigated and it was
found that cooperation between the
US and PU for CHE was possible19. The
principle of cooperation was accepted
by the PU for CHE on 1995-01-31.
The principle of interuniversity
cooperation was also attractive to the
University of Pretoria, who joined the
consortium.
The mechanism of cooperation
between the three different universities
required a lot of negotiations. Initially,
it was considered to conduct the
cooperation in the form of a business
trust, Unitrust. It was, however, realised
that the VAT implications of a business
trust would make such an arrangement
not financially rewarding. It was decided
to use an Interuniversity Information
Technology Committee (Unikom) as
entity that would coordinate systems
development between the universities.
Unitrust served as its marketing wing.
The PU for CHE donated R500 000
to the University of Stellenbosch in
recognition of the development costs
that the latter made to the systems
that would, in future, be shared by the
institutions20.
A project plan was developed for the
implementation of the Unikom systems
at the PU for CHE. The ITM Director was
asked to act as Project Manager and it
was decided to relieve him of some of
his regular responsibilities by having
the managers of the ITM dividions
rotate as acting deputy directors of
ITM for periods of approximately three
months each21.
Initially, both the student systems
and the financial core systems were
scheduled for implementation at the
beginning of 1996. By October 1995,
it became clear that the Unikom
15 ITM Coordination Committee, Minutes, Art 4.1.2, 3 July 1991 16 ITM Coordination Committee, Minutes, Art 3, 22 August 1991 17 ITM Coordination Committee, Minutes, Art 3.3.1, 2 December 1991 18 ITM Annual Report, Art 3.1.2.1, 1994
19 ITM Annual Report, Art 3.1.2.3, 1994 20 Letter from Dr Cobus van der Walt to Prof Kobus Meij, 16 September 1996 21 ITM Annual Report, Art 3.1, 1995
102 103
development team was behind
schedule and that it would not be
possible to implement the student
system in January 1996. Consequently,
the implementation was delayed to
July 1996. The rest of the systems,
excluding the manpower system, were
all implemented during 1996. The
manpower system was successfully
implemented in 1998.
The Unikom systems were developed
for the Adabas database system that
could run on a variety of computer
platforms. With the advent of powerful
processors in a smaller format, it was
no longer necessary for a university
the size of the PU for CHE to use
mainframe equipment. It was decided
to use a cluster of two DEC Alpha
computers for the Unikom systems.
This had the advantage that it was the
same generation of computers that
was in use at the US, who had expertise
in this field. Using the same computer
platforms had the benefit of complete
interchangeability of software, sharing
of equipment, etc. The DEC Alpha
computers were installed in July 1995
and commissioned in October 1995.
The computers were stable and
rendered good service22.
The University’s large business systems
that formed part of the Unikom systems
were no longer dependent on the
mainframe computer and the smaller
business systems were migrated
to the client/server environment.
Word processing and SAS was
earlier migrated to PCs and scientific
processing had already been migrated
to the RS/6000s in 1990/1991. It was,
therefore, possible to phase out the
mainframe computers of the University
and decommission them in November
1966. Historical data was moved to
the University of Pretoria’s mainframe
computer beforehand to ensure safe
backup23, if ever needed.
In January 1997, the Unikom system was
used for the first time during student
registration. The system was first tested
with the smaller umber of students
at the Vaal Triangle campus, and
problems were addressed. Thereafter
it was successfully implemented at the
Potchefstroom campus24.
In 1998, Version 2 of the Unikom
systems was implemented after a
new Compaq Alphaserver 4100 was
commissioned to handle the increased
load. This made the old DEC Alpha
computers available as the testing
platform for system development and
to run adaptability tests for millennium
compatibility.
The new Compaq computer had
adequate computing power to easily
handle the peak time during the 2008
student registration period and, for
the first time, it was possible for other
computer users to continue with
their tasks while registration was in
progress25.
The University of Stellenbosch
unexpectedly announced in May 1999
that they wanted to terminate the
four-year-old Unikom cooperation
agreement. Although the PU for CHE
was unsatisfied with many aspects
of the cooperation, especially the
management by the US of the
development group and quality control,
the termination came as a surprise and
required an urgent alternative. The
University did, however, continue to
have the right of using and expanding
the Unikom systems, and also obtained
its source code.
In the preceding years, the PU for CHE
rationed its staff complement in ITM
Information Systems division, with
the result that there was no longer
sufficient human resources to support
systems of this extent.
22 ITM Annual Report, 1996 23 ITM Annual Report, Art 2.6, 1996
Figure 12 Mr Ernest van Niekerk with one of the DEC Alpha computers on which the Unikom systems operated. Mr Van Niekerk was the ITM Information Systems manager for a long time. (Source: NWU archive)
24 ITM Annual Report, 1997 25 ITM Annual Report, Art 3.2, 1998
Figure 13 Mr Johan Greyling with one of the IBM 3380 disk drives during the dismantling of the mainframe computer in 1997. The IBM 3380 disk drives was an important component of the mainframe environment and had a capacity of 2.5 gigabytes. Mr Greyling was a key figure in ITM Information Systems for a long time. (Source: NWU archive)
104 105
Business system support
in cooperation with IT3
Educational Systems
The University’s dilemma with the
termination of the Unikom agreement
was used as an opportunity to
significantly improve the development
and support of the University’s business
systems. In this regard, further system
development and maintenance was
contracted to a new company, IT3
Educational Systems.
IT3 was a group of IT companies
mainly based in the Western Cape.
The Managing Director of IT3, Mr
Don Hobson, also functioned as a
consultant in the system development
group of both Unikom and the US.
With the dismantling of the Unikom
agreement, several of the Unikom
contractors indicated that they were
willing to provide further support
services to the PU for CHE, albeit in
a different capacity. A new company
was established in the IT3 group in
August 1999, namely IT3 Educational
Systems (IT3:ES), in which the PU held a
minority share and could nominate two
directors on the board of five. The first
board meeting of IT3:ES took place on
1999-11-18 in Durbanville26. The PU for
CHE’s directors on the board were the
Rector, Prof CJ Reinecke and the ITM
Director, Dr AJ van der Walt. By the
end of 1999, the company employed
approximately 18 contractors to
support and further develop the
University’s Unikom systems. Almost
all the contractors were experienced
in the Unikom systems and previously
functioned as contractors to Unikom27.
The PU for CHE was impressed by the
dramatic improvement in the Unikom
system’s management and the resulting
quality improvement. The end-users
were also happy with IT3 ES’s service.
A lot of work was conducted in the
Unikom ranks to ensure that the
Unikom Version 3 release of the
business systems would be millennium
compatible. It was therefore imperative
that Version 3 be implemented
before the end of 1999. The sudden
termination of the Unikom cooperation
agreement at the end of July 1999, as
previously described, put the PU for
CHE under immense pressure. The
newly-formed IT3:ES did, however,
through close cooperation with the
ITM’s Information Systems division,
ensure that Version 3 could be
implemented in 1999. It was testified by
the personnel of the University that this
was the easiest implementation of any
of the Unikom versions28, a testimony
to the high-quality service that the
University consistently received from
IT3 ES.
IT3 ES’s development of new
functionality in the University’s
business systems was done efficiently.
Version 4 made provision for the
altered requirements of reporting
to the Government. Version 5 was
implemented in July 2000 because
the University’s academic structure had
changed and provision hat to be made
for the establishment of academic
schools and focus areas29. In 2000, three
new versions of the business systems
were developed and implemented to
address changing needs.
By the year 2000, it was evident that
business systems were not being
developed quickly enough to keep
track with the University’s business
needs. This was, in particular, ascribed
to the ineffective Natural development
environment of the Adabas database
system. In cooperation with the
Gartner Group, an investigation was
launched into alternative development
environments and it was recommended
to the Management Committee that
the Forté development environment be
purchased. This recommendation was,
eventually, approved by the Rector.
The change required that the IT3 ES
and University’s development staff be
trained in the use of Forté. This process
was completed by October 2000.
The implementation of the Forté
development environment enabled
the University to think differently
about system development. In 2001, it
was decided to replace the business
systems with a new development,
Epuk. This was an extensive and
comprehensive project that would take
a few years and provide the University
with high-quality business systems.
In 2001, the University established a
company, Glob-Ed Systems (Pty) Ltd,
wherein ownership of the new business
systems would reside30. This company
would also handle the marketing of
these systems. A decision was made
to refer to the business and education
training systems as the Varsité systems
for marketing purposes.
By the time that the PU for CHE
became part of the multi-campus
North-West University, it had a good
set of business systems that would be
further developed according to
leading standards.
26 IT3 Educational Systems, Board Meeting, Minutes, 18 November 1999 27 ITM Annual Report, Art 3.4, 1999 28 ITM, Annual Report, Art 3.5, 1999
29 ITM, Report to Audit Committee, Art 3.1.1, Mei 2000 30 Glob-Ed Systems (Edms) Bpk, Minutes of first board meeting, 2001-07-17
106 107
TEACHING AND LEARNING
MANAGEMENT SYSTEMS
In the 1990s, ITM promoted the
implementation of electronic learning
methods (e-learning) by the University.
ITM tried to motivate the Management
Committee to implement e-learning
as a general teaching and learning
method. ITM conducted extensive
background research in this regard and
often discussed the matter during the
Management Committee’s visits to the
ITM.
In 1996, the University decided to
offer a B-degree in Business
Administration (BBA) telematically for
distant students31. For this purpose
the department of Telematic Learning
Systems (TLS) was established with
Prof Lou van Wyk as Director. These
Figure 14 The task team for network- based learning and teaching platforms during a visit to the University of Michigan. From left to right: Prof Paul du Plessis, an IBM representative that travelled with the group, Dr Cobus van der Walt and Prof Albert Viljoen. (Source: Author’s photo collection)
developments departed from the
practice of using only contact sessions
as a teaching and learning method. TLS
initially used Lotus Notes as a teaching-
learning management system, but it
had defects and was later replaced with
a product developed internally.
One of the basic needs for the
implementation of e-learning was the
availability of a teaching and learning
management system. This system
is used to grant local and distance
students telematic access to learning
material, communication with lecturers
and fellow students, and access to
relevant library and administrative
systems. Attempts were made from
1996 onwards to develop such a system.
The first project was PUKWESS (PUK
Web Student Support). Mrs Bessie van
der Walt from the Ferdinand Postma
Library, was seconded (through the
Vice Rector’s mediation) to ITM User
Training for a period of eight months
to help with the system’s development
because she was experienced in
web-based systems development.
This proved very successful and she
managed to complete the first web-
based registration of a prospective
student32.
A task team that consisted of the
Vice Rector, Prof AJ Viljoen; Prof SJP
du Plessis, Director of the Bureau for
Academic Support Services; and Dr AJ
van der Walt, Director of ITM, conducted
visits to the USA and Canada from 26
May 1997 to 11 June 1997 to study
network-based learning and student
support systems33. It became clear to
them that the University urgently had
to develop a strategy to initiate and
implement network-based learning.
In response to the aforementioned,
the development of a teaching and
learning platform garnered more
attention and, subsequently, a study
group on student support platforms
was established. This group decided
on the name FLIP (Flexi-learn Internet
Platform) on 28 October 1997 for
the system of the PU for CHE34. It
was decided to use a development
package, SERF, from the University of
Delaware to further develop FLIP35.
SERF was used by both the PU for CHE
and the University of Stellenbosch as
development package.
In 1998, the members of Unikom
nominated a task team to make a
recommendation on a teaching and
learning system. This group, under
the leadership of Ms Karen Lazenby
from the University of Pretoria,
recommended on 13 November
1998 that WebCT be selected by the
Unikom partners as teaching and
learning management system36. The
FLIP task team, under the leadership
of Attie Juyn, the Deputy Director of
ITM, recommended on 16 July 1999,
however, that the PU for CHE rather
continue with the development of its
own FLIP, since many shortcomings
of WebCT had come to light in the
interim.
In 1999, as explained above, the
development and support of the
University’s business systems moved
from Unikom to IT3:ES. IT3:ES
announced that they wanted to
develop a new and comprehensive
teaching and learning platform,
Alexander. This platform would
31 CJ Reinecke, Presentation to University Board: Vestiging van ‘n Telematiese Leersisteem by die PU vir CHO, 1996-04-12
32 Transcription of Author’s interview with mnr Ernest van Niekerk, 15 September 2009 33 AJ Viljoen, SJP du Plessis and AJ van der Walt, Report on the visit to the USA and
Canada to determine the implications of the changing learning environment for the PU for CHE, IT-geskiedenis\Winword\Verslag oor besoek aan VSA en Kanada.doc, 20 June 1997
34 Study group on the student support platform, Agenda, Art 5, 28 October 1997 35 KTask team on Technology and Flexible Learning, Report, Art 4, 25 February 199 36 Karen Lazenby, Report On The Evaluation Of Web-Based Teaching\Learning
Management Systems By The Universities Of Stellenbosch, Potchefstroom and Pretoria, Art 6, 13 November 1998
108 109
completely integrate with the
University’s business systems. The
suggestion was accepted by the
University after concepts thereof were
also evaluated by Prof Casper Schutte,
the former Chief Executive Director of
Science, Technology and Informatics at
the University of South Africa37.
The first version of Alexander was
implemented in the second semester
of 2000 and successfully used for the
presentation of the Pharmacology
Honours programme38.
Alexander was based on the Microsoft
development environment, and there
was positive interaction with Ms Leanne
Steer from Microsoft South Africa
regarding Alexander. She indicated
that Microsoft was in the process of
Figure 15 The signing of the Memorandum of Understanding between the PU for CHE and the Learning Lab of the University of Wolverhampton on 15 November 2000 by Dr Cobus van der Walt and Prof Steve Molyneux, respectively from the two institutions. (Source: Author’s photo collection)
developing their own teaching and
learning platform, Microsoft Learning
Technology (MLT). This information
was very important for Alexander’s
further development, since it would
be meaningless to compete with
Microsoft. It would be preferable to
rather enrich MLT.
On 23 June 2000, Leanne Steer
facilitated a meeting in Cape Town
between the PU for CHE IT Director,
IT3:ES and prof Steve Molyneux from
the Learning Lab at the University
of Wolverhampton in England. The
Learning Lab was in the process of
developing a teaching and learning
management programme named Wolf.
Several similarities were identified
between Alexander and Wolf and it
was decided in principle that
representatives of IT3 ES and the PU
for CHE would conduct a visit to the
Learning Lab shortly thereafter. This
visit took place from 24-28 July 200039.
Molyneux and his staff’s approach
appealed to the representatives,
and was similar to their approach
to Microsoft’s MLT. The Rector, Prof
Carools Reinecke, conducted his
own visit to Molyneux and his team in
August 200040 with the intent to foster
greater cooperation. It was decided to
attempt the establishment of Learning
Lab at the PU for CHE, like that of
Molyneux. As a result, a Memorandum
of Understanding was signed on 2000-
11-15 by Prof Steve Molyneux and Dr
Cobus van der Walt at the University of
Wolverhampton41.
The plans for Alexander’s linking to
Microsoft’s MLT was foiled when
Microsoft announced in December
2000 that they would were abondoning
MLT’s development42. The development
of Alexander did, however, continue.
Alexander version 2.1 was ready by the
end of 2001 and was used to present
the compulsory module, RINL 111
(Computer and Information Skills)
to all (approximately 3 000) first-year
students in 2002. In their evaluation of
Alexander, the Bureau for Academic
Support Services reacted positively to
the characteristics of Alexander and
stated that it was the best teaching and
learning system available at the time43.
With the transfer of the marketing of
the University’s business systems to
Glob-Ed Systems (Pty) Ltd, the name
Alexander was changed to Varsité
Learning Management System in
200244.
Varsité Learning Management System
was further developed and extensively
utilised by the University. The
development of an in-house teaching
and learning management system
proved to be a wise and cost-effective
decision, since its cost was much less
than that of a good commercial system.
37 CJH Schutte, Report on the Alexander project, 29 April 2000 38 ITM Information systems, Annual Report, Art 2.6.2, 2000 39 Cobus vd Walt, Internal Report, Important impressions of visit to England, 24-28
July 2000, 1 August 2000 40 ITM , Management Meeting, Minutes, Art 11.5, 2 August 2000
41 Cobus van der Walt, Report, Visit to the University of Wolverhampton, Art 3.1, 15 November 2000
42 Telephone conversation between Leanne Steer and Cobus van der Walt, December 2000
43 Bureau for Academic Support Services, Progress Report: Alexander development project, 26 November 2001
44 Glob-Ed Systems (Pty) Ltd, Board Meeting, Minutes, Art 6.2, 27 February 2002
110 111
THE Y2K PROJECT
Towards the end of the twentieth
century, the IT industry became aware of
potential problems in many computer
systems, since only two numbers were
used to store the year number instead
of the entire year comprising four
numbers. This detrimental practice
originated from an era when computer
memories and disc space was very
expensive and this practice could save
space by just using two characters for
the year number. As a result, a lot of
computer and management systems
could experience problems at the turn
of the millennium at the end of 1999.
Many institutions took the necessary
precautions to overcome this problem,
which necessitating editing of the
relevant programs. These programming
updates, in turn, had an impact on
equipment requirements and, in many
cases, required the purchase of new
equipment. The risk also existed of
interruptions to the national power grid
and, as such, provision had to be made
for emergency electricity supplies. At
many organisations, including the PU
for CHE, projects that had to include
senior management, were launched for
the year 2000 transition.
The first mention of the Y2K problem
was made in a newsletter to the ITM
personnel after a Management
meeting on 29 November 1995. It
was mentioned that Cobus van der
Walt, the ITM Director, attended a
symposium of the Gartner Group, a
well-known advisory organisation in
the field of information technology
held in France. At this symposium,
he became aware of the attention at
a high-level was being given to the
Y2K problem. ITM prepared itself to
compile a technology vision for Y2K,
which would allow it to plan for the year
2000 from the correct framework. This
vision was completed in 199645.
One of the management goals of ITM
in 1997 was determining the impact
of the year 2000 and planning for
it accordingly. Mrs A Oosthuyzen,
Manager of ITM Vaal Triangle Campus,
was nominated to lead a study group
in this regard and make the necessary
recommendations. The ITM Advisory
Committee meeting of 1997 warned
that the year 2000 should be given
high priority46. During the Management
Committee’s annual visit to ITM in
1997, the Y2K problem was brought
to their attention47. This discussion
was repeated during the 1998 and 1999
meetings. At a joint meeting of the
ITM Coordination Committee and the
Monitor Committee of Administrative
Computer Systems on 30 October 1997,
it was decided to recommend that a
University Committee be nominated
to plan for the implications of the
millennium transition48. The biggest
risks that the aforementioned meeting
identified included the following::
• Problems with the official University
systems
• Important research and other
academic systems that were not
centrally managed
• Individual work stations that were
not adequately tested or replaced
in time
• Hidden microprocessors, e.g. in
laboratory equipment, elevators,
etc., which were not Y2K
compatible
• University service providers that
were not Y2K compatible
• A large demand could develop for
IT experts, especially closer to the
turn of the millennium. This could
put the University in a difficult
position regarding retaining ITM
expert staff.
A warning was issued that the Y2K
could have significant cost implications
for the University.
In 1998, a University-wide Y2K project
was launched under a high-level
steering committee. Mrs A Oosthuyzen
was appointed project leader.
In the interim, Unikom thoroughly
investigated the Y2K compatibility of
their systems, an action that would
continue until the dissolution of the
partnership in 1999. A computer
programme, Insight 2000, was
purchased by Unikom to identify risks,
and a new Y2K-compatible Version
(Version 3) of the Unikom system was
developed.
All ITM’s systems were reviewed for
possible Y2K problems that had to be
corrected.
The increased equipment demands
of the Y2K-compatible network and
anti-virus programmes, necessitated
the replacement of incompatible
computers connected to PUKNET.
A total of 345 workstations had to be
replaced and the memory of 400
workstations had to be upgraded.
A cost of R2,4 million was budgeted
for49. Funds were also made available
45 Cobus van der Walt, Technological vision for 2000, 5 September 1996 46 ITM Advisory Committee, Minutes, Art 6.6, 2 April 1997 47 Visit from Management Committee to ITM, agenda, Art 5.2.3, 12 August 1997
48 ITM, Overview Report to the Audit Committtee, Art 3.1, 6 November 1997 49 Steering Committee Year 2000, Minutes, Art 5.4, 26 August 1999
112 113
for the purchase of a motor generator
for ITM’s machine room.
As a precaution, the Y2K Steering
Committee decided that no ITM staff
member would be allowed to take
leave during January and the first half
of February 200050.
The Y2K Steering Committee did
thorough work and ensured that most
people at the University were aware of
the potential Y2K problems and their
precautions.
On 2000-01-01, a special ITM
management meeting was scheduled
to determine whether the millennium
transition was progressing smoothly51.
The meeting was also attended by
the Vice Rector, Prof AJ Viljoen; the
Registrar, Prof CFC van der Walt; and
the Registrar of Finance, Prof IJ Rost.
It was gratefully noted that no big
problems had been observed and
that all business systems that could
be tested at that stage, functioned
correctly. No serious problems were
detected later either. At the end of
the meeting, Prof Viljoen thanked ITM
for managing a smooth millennium
transition at the University.
The planning for the millennium
transition under the leadership of Mrs
Aldine Oosthuyzen was, therefore, very
successful.
THE EXPANSION OF THE
PC LABORATORIES
Software legalisation
One of ITM’s inherited problems was
the fact that very little of the software
in use in the PC laboratories were
legal. With the end of sanctions
against South Africa in the early 1990s
and intensified copyright legislation,
it became necessary to control the
legality of software. Unfortunately,
the lack of control during the sanction
years resulted in many illegal copies of
software being used at the University
and that staff members often engaged
in unethical and illegal behaviour in
this regard. It could have been a big
embarrassment for the University if
any legal action was taken against it
as a result of this. The legalisation of
software had a big cost-implication for
the University, which is why it was not
easy to rectify the problem during the
early budget-restricted years.
In 1992, a two-year plan was launched
to legalise the software in use in the
central training labs. Half of the cost
of approximately R200 000 (about R900
000 in 2017 ZAR value) was obtained
from the 1992 budget and the rest was
purchased early in 1993 from that year’s
budget. This legalised the standard
software that ran on PUKNET, the local
area network of the University, and the
central PC labs under ITM control52.
University personnel were informed of
the requirement to only use legalised
software.
From 1993, provisions were made to
ensure that ITM had the necessary
software licenses to ensure that
simultaneous users of the central IT
facilities were within legal bounds. This
system worked well until 1997, when
Microsoft no longer sold simultaneous
licenses for their software, and ITM
had to buy individual packages.
This resulted in negotiations with
Microsoft South Africa’s Leanne Steer,
the Microsoft representative at the
University, with whom the University
had a good relationship. ITM obtained
campus licenses from Microsoft for
their products and operating systems
that were in use. The annual costs
amounted to R200 000 (about R550
000 in 2017 ZAR value)53. This included
departmental computer use and home
use for staff members which were on
the payroll of the University. Although
it was substantial expenditure, this
ensured that the University kept its staff
members and students within legal
bounds regarding Microsoft software
use. This arrangement was still in effect
by the time that this document was
published.
50 Steering Committee Year 2000, Progress with high-risk areas, Art 1.15, 28 April 1999 51 ITM, Management Meeting, Minutes, 2000-01-01
52 ITM, Annual Report, Art 3.6, 1992 53 ITM, Management Meeting, Minutes, Art 7.2, 24 February 1989
114 115
The expansion and
upgrading of the PC
laboratories
The continuous demand for more and
bigger PC laboratories was a result
of the intense use of the laboratories
by certain courses. As a result of the
pressure on ITM’s capital budget in the
first few years (pressure that resulted
from overspending in 1990 that had
to be paid back over three years),
no significant PC room expansion
could be undertaken. Thereafter,
there was consistent expansion and
upgrading of the PC labs. The PC labs
benefitted greatly from the upgrading
of computers to prevent Y2K problems.
Many of the labs were upgraded
to Pentium processors during this
period. This resulted in a significant
improvement (Figure 16).
In 2001, the PC laboratories in the
Library were expanded and 144 new
Pentium 4 PCs were provided. At
that time, the total number of PCs in
laboratories under ITM’s management
was 409. In addition, 286 more PCs
were distributed across 10 laboratories
in different Departments or Faculties
THE DEVELOPMENT
OF PUKNET
Initial implementation of
PUKNET
The coordinated development of
PUKNET commenced in 1990, as
described in Chapter 4. The
Potchefstroom campus network was
initially developed with a back-bone
connection in ring form between
the most important buildings, with
network servers located at key points
on campus. Interior cabling was initially
based on the token ring standard, but
was later done according to the 10
Base-T standard. On the Vaal Triangle
Campus, the network initially started
as a small network between the earlier
engineering buildings. The interior
cabling there was based on the
Ethernet Thinnet standard.
PUKNET quickly developed into a key
element of the University’s infrastructure
and many people depended on it for
daily tasks. The further expansion and Figure 16 The growth in the number and quality of workstations in the PC laboratories. After the upgrading in 1999 to prevent Y2K problems, the increase in computing power is evident. (Source: ITB Annual Report, 1999)
stabilisation thereof was, therefore, a
high priority.
116 117
Further development of
PUKNET
With time, PUKNET expanded
almost exponentially. By 1994, the
Potchefstroom campus had 1600
microcomputers that were connected
to the network. In 1995, PUKNET
was also expanded to buildings
that previously belonged to the
Potchefstroom Education College.
The Vaal Triangle Campus network also
developed rapidly. All the important
buildings were connected by 1994 and
at that time most staff members had
network connected microcomputers.
All main frame terminals at the Vaal
Triangle Campus were phased out
during 1994.
By 2001, there were 2 989 network
ports at the Potchefstroom campus
and 272 at the Vaal Triangle campus.
The increase in the number of network
ports during the 1990s is depicted in
the following figure (Figure 17).
From a relatively small network in
1991, PUKNET expanded into a large,
comprehensive network by the end of
the 1990s.
Cabling standard
The mixed cabl ing standards
undermined the network’s reliability
and made maintenance difficult. A
programme was quickly launched to do
all cabling according to the 10 Base-T
standard. By 1995, both campuses
adhered to this standard.
The rationalisation of the connection
standards resulted in the networking
being more manageable, new
connections being easier to install and
problem identification being easier.
Topology and technology
The increase in network traffic at the
Potchefstroom campus made the
backbone topology that was initially
installed ineffective. A comprehensive
investigation was launched into
network topology and related
equipment was installed. In 1994 it
was decided to make use of a so-called
collapsed backbone. This implied that
all connections to network servers
were combined at a central point at a
fast router, a Wellfleet BCN. As part of
this project, all network servers were
moved to the machine room where
maintenance, a stable power supply
and environment were available.
Later on, ITM Networks developed
a strategy for further upgrading
by the Management Committee in June
1998 and the cost was approximately
R4,6 million over a period of three
years. For this purpose, Fore/Marconi
equipment was used and implemented
from 1999. This gave PUKNET a very
fast backbone and ensured high-speed
connectivity to the large number of
network users.
Network operational
system and application
software
At the start of the PUKNET
development, it was decided to use
Novell Netware as network operating
system. It appeared to be an excellent
choice and during 1996, the system was
upgraded to Netware 4.1.
Initially, the Da Vince email system
was implemented, but the limitations
thereof resulted in the implementation
of Groupwise, also a Novell product, in
1996. This software delivered excellent
service for a long time
Storage space
By 1995 the available storage space
on the file servers of PUKNET was 48
gigabytes on magnetic drives and 20
gigabytes on optical drives. This
increased in 1996 to almost 100
gigabytes in total. Later numbers are Figure 17 The growth in the number of network points at the Potchefstroom and Vaal Triangle campuses in the 1990s. (Source: ITM Annual Report, 2001)
of PUKNET through very fast ATM
technology. This strategy was approved
not known.
118 119
Internet access
Initially, internet access to the University
was provided by the Uninet consortium
of South Africa’s public universities.
During 1991 only a dial-up connection
was available for international
connectivity and it could only be used
for email. From 1992, Uninet used a
dedicated line of 4 800 bps for the
internet connection. This was increased
to 9 600 bps in 1993 and to 64 kbps in
1994. With time, the University started
using an additional dedicated line
to SAIX. In 2000 this was a 512 kbps
connection54.
The following figure depicts the
significant increase in bandwidth from
1991 to 2002.
Student access to email and
internet
During 1995 network connections were
made to a computer in each student
residence, and a trial run was conducted
to install network connections in all the
rooms of the Patria and Wag-’n-Bietjie
student residences.
Access to facilities like email on the
network for students was provided in
cooperation with the student
organisation, Sleutelbord, that was
established in 1995. These services
were significantly expanded from 1997
onwards, as additional financing was
obtained from an additional R25 levy
on student registration fees.
LOAD SHEDDING SYSTEM
In large organisations, the electricity bill
depends on both the amount of kWh
used and the peak amount of electrical
energy used in a period of 15 minutes
in a given month, reported in kVA. It
is therefore important to minimise
the maximum KVA demand as far as
possible. In 1990, at the Potchefstroom
campus a system was installed which
managed KVA demand by switching
off geysers during peak time. This
system was initially implemented by
Electronic Services and from 1991 was
the responsibility of ITM Networks.
The load control system resulted in a
big saving on the University’s electricity
bill. In the next figure, for example,
the kVA demand from 1985 to 1996 is
depicted in a column graph, with the
solid lines indicating the electricity
costs per year and the projected
electricity costs without load shedding
by a dotted line. It is clear that, when
load control was implemented in 1990,
the KVA demand was clearly lowered,
as did the electricity costs. By 1996 the
load shedding saved approximately
R1,5 million per year.
After 1996, the actual electricity costs
started increasing at the same gradient
as before 1990, since all the loads that
could be controlled was already part of
the load shedding system.
Figure 18 The increase in internet bandwidth and the associated costs from 1991 to 2002. (Source: Annual Report, ITM Networks 2001, Art 3.1.4)
54 ITM networks, Annual Report, Art 3.1.6, 1996
Figure 19 The trend of KVA demand (column graph), the actual electricity costs (solid line) and the projected electricity costs without load shedding (dotted line) between 1985 and 1996. (Source: Annual Report ITM Networks 1996, art 3.4)
120 121
THE IBM SP RESEARCH
COMPUTER
There was a great need under
certain researchers, especially in the
Departments of Physics, Statistics and
Computer Science and Information
Technology for powerful computers
for their very computing-intensive
research. During 1996 an interest
group of these researchers and staff
from ITM negotiated with IBM about
the provision of such a powerful
research computer to the University.
The negotiations were successful and
the computer was ordered by the Vice
Rector, Prof Albert Viljoen, on 1996-12-
0655.
The IBM RS/6000 SP computer with
parallel processors was delivered in
January 1997 and was made available
for use in February 1997. The purchase
of the computer was made possible
by a grant from the SUR (Shared
Universities Research) programme of
IBM. The grant enabled the University
to purchase a computer with a list price
of R3 500 000 for R1 000 00056.
The IBM SP was powerful enough to
allow scientists at other universities
to also evaluate the system. In
doing so, the PU for CHE tried to
influence other universities to join a
partnership of universities by financially
contributing to the upgrading of
the SP. This would have allowed
the development of a true national
academic research computer centre.
The University of the Witwatersrand’s
Department of Chemistry was the first
to test the SP, followed by the Centre
for Water Resources Research (in
Pietermaritzburg) and the University
of Stellenbosch. At the PU for CHE,
these facilities were mainly used by the
departments of Physics, Statistics and
Computer Sciences.
Unfortunately, additional funds for the
upgrading of the IBM SP did not
realise and, in time, other computer
platforms overtook it. It did, however,
provide very good service for quite
some time to the researchers at the
PU for CHE. Prof Marius Potgieter
from the Department of Physics was
an intensive user of the IBM SP and
highly praised its ability to be used in
advanced research57 projects.
THE RELATIONSHIP
BETWEEN THE PU
FOR CHE AND IBM
The PU for CHE’s first computer was
provided by IBM South Africa and
installed in July 1967. This was the
beginning of a long partnership
between the University and IBM
South Africa, as well as its successor,
ISM, after IBM’s withdrawal from South
Africa in 1986. For many years IBM
dominated the mainframe environment
and was a source of expertise in various
fields. All the University’s mainframe
computers (the IBM 1130, 370/125,
4331, 4341 and 4380) were supplied
by IBM. Later, when RISC processors
were developed, the PU for CHE was
the first South African University to
purchase a RS/6000 computer. Later
on, in the 1990s, IBM South Africa
made a special offer to supply the IBM
SP parallel processing computer to the
University, as described above.
IBM South Africa and ISM made several
donations to the University, the earliest
of which are described in Chapter 2.
In the late 1980s, an IBM 7350 image
processer was loaned to the University
(also as per a previous Chapter).
Figure 20 The cover page of the 25-year anniversary festivities of the IBM and PU for CHE partnership (Source: Author’s document collection)
In the University and IBM South Africa/
ISM partnership, the latter’s university
representatives played a significant role
at Computer Services/ITM. Mention
has already been made of Mr Berthol
Theiland’s role in the 1980s, and in
the first half of the 1990s, Ms Mariet
Venter provided excellent support.
On 24 August 1992, both sides were
delighted to commemorate the 25-year
partnership during a festive occasion.
55 Vice Rector of the PU for CHE’s letter to IBM South Africa, 6 December 1996 56 ITM, Annual Report, Art 3.1, 1997
57 Transcription of the Author’s interview with Prof Marius Potgieter, 20 October 2010
122 123
The celebration took the form of a
dinner in the reception room of the
synod hall of the Reformed Churches
in Potchefstroom. At this event,
Prof JF de Beer (the first head of
Computer Services), was the master
of ceremonies, and Mr JF Clarke (Post
Office and Telkom CEO and former
IBM South Africa CEO) was the guest
speaker. The prayer before the meal
was done by Prof WP Robbertse, the
former Dean of Natural Sciences, who
strongly advocated for the purchase
of the first computer and for the
establishment of Computer Science
as a University subject. Mr Wiid du
Toit, the IBM salesman of 1967, and
Mrs Klaasje Benadie, the first full-time
employee of the Computer Centre,
were also in attendance. At this event,
Mr Malcolm Taylor, General Manager of
Smaller Companies in ISM, presented
a commemorative wall clock to the
Director of Information Technology
and Management.
For a long time, IBM was the leader in
the computer world at the time when
the mainframe computer was still king
of computers in big organisations.
With the general downsizing of
computers, amongst others as a result
of the budget pressures of the 1990s
in the USA, IBM’s role in the world of
powerful processors and client-server
processing diminished, resulting in IBM
losing its leadership status. As a result,
IBM’s role at the PU for CHE became
less important.
Figure 21 21 The commemorative clock that IBM South Africa handed presented to Information Technology and Management at the 25-year celebration of IBM computers at the PU for CHE on 24 Augustus 1992. (Source: Photo collection of Mr Boeta Pretorius)
Figure 22 At the commemoration of the 25-year IBM and PU for CHE partnership on 1992-08-24, a photo was taken of ITM management. From left to right: Mr Ernest van Niekerk, Manager of Information Services, Dr Cobus van der Walt, Director of ITM, Mr Jan van Rooy, Manager of Operating Systems, Mr Piet Vergeer, Head of Administrative Support, Mrs Aldine Oosthuyzen, Manager of ITM at the Vaal Triangle Campus, Mr Boeta Pretorius, Manager of User Training and Mr Ernst Blignaut, Manager of Electronic Services. (Source: Photo collection of Mr Ernest van Niekerk)
Figure 23 Some of the guests who a t tended the 25 - year commemoration of the IBM/ISM and University partnership. From left to right: Ms Mariet Venter, Sales Representative of ISM, Mr Malcolm Taylor, General Manager of Smaller Companies at ISM, Mrs and Mr JF Clarke, CEO of the Post Office and Telkom, Prof Carools Reinecke, Rector, Mrs Bessie van der Walt and Dr Cobus van der Walt, Director of ITM. (Source: Photo collection of the NWU Archive)
124 125
Figure 24 TThe IT-heads/directors who attended the 25-year commemoration: Prof Giel Hattingh (1981-1985), Dr Philip Welman (1986-1990), Prof Hannes de Beer (1967 - 1978) and Dr Cobus van der Walt (1990 - 2000). Mr Johan Smith (1978 - 1980) could unfortunately not attend the event. (Source: Photo collection of the NWU Archive)
AFRIKAANS
SPELLCHECKER
As mentioned earlier, an Afrikaans
spell checker that was developed by
IBM SA was supplied to the University
for word processing in the mainframe
and DisplayWrite environments. With
the general use of Microsoft Word on
PUKNET and campus PCs, the
need developed for an Afrikaans spell
checker for MS Word.
ITM User Training identified this
gap and in 1994 did the necessary
development work for a Microsoft
Word compatible spell checker, and
they started by compiling an Afrikaans
word list. This developed in a successful
commercial product that sold 3 922
copies in 199558. The product was often
updated for new versions of MS Word.
It was later commercially marketed
outside of the University and in 2001,
13 807 commercial licenses59 were sold.
58 ITM Annual Report, 1995, Art 32 59 ITB Customer Services, Annual Report, 2001, Art 4.1
126 127
STYLE TEMPLATES
A stylesheet contains a series of
instructions for a word processor,
like Microsoft Word, to standardise
document layout and ensure
consistency. This is used in many large
organisations and forms part of the
corporate identity of the organisation.
It also simplifies document assembling
from subdocuments originating from
different sources.
ITM emphasised the advantages of
stylesheets early on to the Vice Rector,
Prof Albert Viljoen. Prof Viljoen was
supportive of the idea, since his office
encountered many problems with
the compilation of annual and other
comprehensive reports that originated
from diferent offices. The Vice Rector
appointed a task team early in 1993 to
compile a page layout standard for the
University. The task team consisted of
representatives from various university
academic and service departments. Dr
Cobus van der Walt, Director of ITM,
served as the convener of the task
team. The task team met seven times
and, on 1994-12-02 provided feedback
to the Vice Rector60.
A set of standard stylesheets were
compiled and approved by the
Management Committee. The
stylesheets were compiled in a way that
would allow for easy implementation.
Some of the advantages of the
stylesheets were:
• A uniform appearance for all
University documents, which was
especially important for documents
compiled from different offices.
• Easy integration of the different
documents from different authors
to form one composite document.
• Time was saved by not having to
experiment with different types of
headings, fonts and font sizes.
• The minimum amount of training
was necessary to enable staff
members to compile neat,
standardised documents.
The Vice Rector made the use of
standard stylesheets compulsory for
all annual reports that were submitted
to him. This assisted greatly in making
the use of the University’s stylesheets
standard practice at the University.
CONCLUSION
After the great developments in the
field of Information Technology in the
1990s, the PU for CHE was, at the end
of its existence as an independent
university, well-placed in terms of its
information technology resources.
From humble beginnings in 1967,
computer, network, knowledge and
learning resources were developed
that resulted in a well-developed
information technology base for the
newly unified North-West University
at the end of 2003 after the merging
of the different campuses of the PU for
CHE.
60 Task team to formulate page layout standard, Report, Convener: Cobus van der
Walt, 2 December 1994
128 129
M
ACKNOWLEDGEMENTS y appreciation to Attie Juyn and
Boeta Pretorius for giving me
the interesting task of documenting
the history of the development of
information technology at the PU for
CHE. In addition, I would like to convey
my gratitude for their assistance in
providing information and other
support.
Mr Jaap Verhoef provided very
important assistance over an
extended period of time in finding
original documents and making them
available to me. Many thanks! I am
also grateful for the NWU Archive who
was extremely helpful in finding and
scanning documents and photographs.
My thanks to the following persons (in
alphabetical order) who were willing
to share their experience through
interviews and otherwise: Mrs Corrie
Breytenbach, Dr Willem Coetzee, Prof
JF de Beer, Prof Giel Hattingh, Mr
Attie Juyn, Dr Tom Larney, Prof Harm
Moraal, Mrs Aldine Oosthuyzen, Prof
Marius Potgieter, Mr Boeta Pretorius,
Mr Peter Schön, Prof Tjaart Steyn, Prof
PH Stoker, Prof Hennie Venter, Mr Jaap
Verhoef, Dr Johan van Loggerenberg,
Mr Ernest van Niekerk, Mrs Alta van Zyl,
and Dr Philip Welman.
Mrs Heinie Venter transcribed the
interviews with great care, something
the author is very appreciative for.
To everyone who made old
photographs available, especially Mrs
Anne Mans who has a very valuable
collection of old photos: thank you. My
thanks are also conveyed to Prof Tjaart
Steyn for the valuable historical items
that he preserved and made available
for photography, as well as the valuable
information about the early years that
he clearly remembers.
Prof Casper Lessing and Dr Tom Larney
was very helpful with the chapter about
the use of computer techniques in the
Ferdinand Postma Library.
I am grateful to Mrs Annatjie Verhoef
who corrected grammatical errors in
the Afrikaans version.
Mrs Yvette Labuschagne, thank you
very much for your organisational work
during this long project.
To my wife, Petro, thank you for giving
me the freedom to spend long hours
working on this history.
This history and the entire computer
development at the PU for CHE was
only possible under the guiding hand
of our Lord.
Cobus van der Walt
December 2017